Showing 1–50 of 111 results for author: Nair, H

Privacy as Social Norm: Systematically Reducing Dysfunctional Privacy Concerns on Social Media

Authors: JaeWon Kim, Soobin Cho, Robert Wolfe, Jishnu Hari Nair, Alexis Hiniker

Abstract: Privacy is essential to fully enjoying the benefits of social media. While fear around privacy risks can sometimes motivate privacy management, the negative impact of such fear, particularly when it is perceived as unaddressable (i.e., "dysfunctional" fear), can significantly harm teen well-being. In a co-design study with 136 participants aged 13-18, we explored how teens can protect their privac… ▽ More Privacy is essential to fully enjoying the benefits of social media. While fear around privacy risks can sometimes motivate privacy management, the negative impact of such fear, particularly when it is perceived as unaddressable (i.e., "dysfunctional" fear), can significantly harm teen well-being. In a co-design study with 136 participants aged 13-18, we explored how teens can protect their privacy without experiencing heightened fear. We identified seven different sources of dysfunctional fear, such as `fear of a hostile environment' and `fear of overstepping privacy norms.' We also evaluated ten designs, co-created with teen participants, that address these fears. Our findings suggest that social media platforms can mitigate dysfunctional fear without compromising privacy by creating a culture where privacy protection is the norm through default privacy-protective features. However, we also found that even the most effective privacy features are not likely to be adopted unless they balance the multifaceted and diverse needs of teens. Individual teens have different needs -- for example, public and private account users have different needs -- and teens often want to enjoy the benefits they get from slightly reducing privacy and widening their social reach. Given these considerations, augmenting default privacy features by allowing them to be toggled on and off will allow individual users to choose their own balance while still maintaining a privacy-focused norm. △ Less

Submitted 21 October, 2024; originally announced October 2024.

In situ Etching of \b{eta}-Ga2O3 using tert-Butyl Chloride in an MOCVD System

Authors: Cameron A. Gorsak, Henry J. Bowman, Katie R. Gann, Kathleen T. Smith, Jacob Steele, Debdeep Jena, Darrell G. Schlom, Huili, Xing, Michael O. Thompson, Hari P. Nair

Abstract: In this study, we investigate in situ etching of \b{eta}-Ga2O3 in a metal-organic chemical vapor deposition (MOCVD) system using tert-Butyl chloride (TBCl). We report the successful etching of both heteroepitaxial (-201)-oriented and homoepitaxial (010)-oriented \b{eta}-Ga2O3 films over a wide range of substrate temperature, TBCl molar flows, and reactor pressures. We identify that the likely etch… ▽ More In this study, we investigate in situ etching of \b{eta}-Ga2O3 in a metal-organic chemical vapor deposition (MOCVD) system using tert-Butyl chloride (TBCl). We report the successful etching of both heteroepitaxial (-201)-oriented and homoepitaxial (010)-oriented \b{eta}-Ga2O3 films over a wide range of substrate temperature, TBCl molar flows, and reactor pressures. We identify that the likely etchant is HCl (g) formed by the pyrolysis of TBCl in the hydrodynamic boundary layer above the substrate. The temperature dependence of the etch rate reveals two distinct regimes characterized by markedly different apparent activation energies. The extracted apparent activation energies suggest that at temperatures below ~800 °C the etch rate is likely limited by desorption of etch products, while at higher substrate temperatures chemisorption of HCl limits the etch rate. The relative etch rates of heteroepitaxial (-201) and homoepitaxial (010) \b{eta}-Ga2O3 were observed to scale by the ratio of the surface energies indicating an anisotropic etch. For (010) homoepitaxial films, relatively smooth post-etch surface morphology was achieved by tuning the etching parameters. △ Less

Submitted 2 August, 2024; originally announced August 2024.

Comments: 12 pages, 8 figures

Three fold Spin valley intertwined Dirac cone in nonmagnetic Weyl Semimetal Pt$_3$Sn$_2$S$_2$ with the Kagome lattice A theoretical DFT perspective

Authors: Ravi Trivedi, Seetha Lakshmy, Heera T. Nair, Alok Shukla, Brahmananda Chakraborty

Abstract: Recent experimental and theoretical research in Pt$_3$Sn$_2$S$_2$ have opened avenues for restoring Dirac dispersion in layered materials. Inspired by this we conducted first-principles DFT calculations unveiling a novel nonmagnetic metal Pt$_3$Sn$_2$S$_2$ with a robust Dirac dispersion. Under strong SOC and high pressure the Fermi surface topology shifts yielding spin valley intertwined Dirac con… ▽ More Recent experimental and theoretical research in Pt$_3$Sn$_2$S$_2$ have opened avenues for restoring Dirac dispersion in layered materials. Inspired by this we conducted first-principles DFT calculations unveiling a novel nonmagnetic metal Pt$_3$Sn$_2$S$_2$ with a robust Dirac dispersion. Under strong SOC and high pressure the Fermi surface topology shifts yielding spin valley intertwined Dirac cones and three fold charge density waves due to inversion symmetry breaking. Thermal conductivity and Seebeck coefficient exhibit an inverse relationship with temperature due to increased phonon collisions attributed to anharmonic phonon phonon interactions via Umklapp processes. Ab initio molecular dynamics and phonon dispersion calculations confirm the dynamic and thermal stability of Pt$_3$Sn$_2$S$_2$ illuminating the origin of its Dirac cone and electronic properties. △ Less

Submitted 20 May, 2024; originally announced May 2024.

Comments: 17 pages, 6 figures, 2 tables [Main Manuscript], and 5 pages, 2 figures [Supplementary material]

NeRTCAM: CAM-Based CMOS Implementation of Reference Frames for Neuromorphic Processors

Authors: Harideep Nair, William Leyman, Agastya Sampath, Quinn Jacobson, John Paul Shen

Abstract: Neuromorphic architectures mimicking biological neural networks have been proposed as a much more efficient alternative to conventional von Neumann architectures for the exploding compute demands of AI workloads. Recent neuroscience theory on intelligence suggests that Cortical Columns (CCs) are the fundamental compute units in the neocortex and intelligence arises from CC's ability to store, pred… ▽ More Neuromorphic architectures mimicking biological neural networks have been proposed as a much more efficient alternative to conventional von Neumann architectures for the exploding compute demands of AI workloads. Recent neuroscience theory on intelligence suggests that Cortical Columns (CCs) are the fundamental compute units in the neocortex and intelligence arises from CC's ability to store, predict and infer information via structured Reference Frames (RFs). Based on this theory, recent works have demonstrated brain-like visual object recognition using software simulation. Our work is the first attempt towards direct CMOS implementation of Reference Frames for building CC-based neuromorphic processors. We propose NeRTCAM (Neuromorphic Reverse Ternary Content Addressable Memory), a CAM-based building block that supports the key operations (store, predict, infer) required to perform inference using RFs. NeRTCAM architecture is presented in detail including its key components. All designs are implemented in SystemVerilog and synthesized in 7nm CMOS, and hardware complexity scaling is evaluated for varying storage sizes. NeRTCAM system for biologically motivated MNIST inference with a storage size of 1024 entries incurs just 0.15 mm^2 area, 400 mW power and 9.18 us critical path latency, demonstrating the feasibility of direct CMOS implementation of CAM-based Reference Frames. △ Less

Submitted 20 May, 2024; originally announced May 2024.

Comments: Accepted and Presented at Neuro-Inspired Computational Elements (NICE) Conference, La Jolla, CA. 2024

Multi-Objective Learning Model Predictive Control

Authors: Siddharth H. Nair, Charlott Vallon, Francesco Borrelli

Abstract: Multi-Objective Learning Model Predictive Control is a novel data-driven control scheme which improves a linear system's closed-loop performance with respect to several convex control objectives over iterations of a repeated task. At each task iteration, collected system data is used to construct terminal components of a Model Predictive Controller. The formulation presented in this paper ensures… ▽ More Multi-Objective Learning Model Predictive Control is a novel data-driven control scheme which improves a linear system's closed-loop performance with respect to several convex control objectives over iterations of a repeated task. At each task iteration, collected system data is used to construct terminal components of a Model Predictive Controller. The formulation presented in this paper ensures that closed-loop control performance improves between successive iterations with respect to each objective. We provide proofs of recursive feasibility and performance improvement, and show that the converged policy is Pareto optimal. Simulation results demonstrate the applicability of the proposed approach. △ Less

Submitted 17 October, 2024; v1 submitted 19 May, 2024; originally announced May 2024.

Realtime Person Identification via Gait Analysis

Authors: Shanmuga Venkatachalam, Harideep Nair, Prabhu Vellaisamy, Yongqi Zhou, Ziad Youssfi, John Paul Shen

Abstract: Each person has a unique gait, i.e., walking style, that can be used as a biometric for personal identification. Recent works have demonstrated effective gait recognition using deep neural networks, however most of these works predominantly focus on classification accuracy rather than model efficiency. In order to perform gait recognition using wearable devices on the edge, it is imperative to dev… ▽ More Each person has a unique gait, i.e., walking style, that can be used as a biometric for personal identification. Recent works have demonstrated effective gait recognition using deep neural networks, however most of these works predominantly focus on classification accuracy rather than model efficiency. In order to perform gait recognition using wearable devices on the edge, it is imperative to develop highly efficient low-power models that can be deployed on to small form-factor devices such as microcontrollers. In this paper, we propose a small CNN model with 4 layers that is very amenable for edge AI deployment and realtime gait recognition. This model was trained on a public gait dataset with 20 classes augmented with data collected by the authors, aggregating to 24 classes in total. Our model achieves 96.7% accuracy and consumes only 5KB RAM with an inferencing time of 70 ms and 125mW power, while running continuous inference on Arduino Nano 33 BLE Sense. We successfully demonstrated realtime identification of the authors with the model running on Arduino, thus underscoring the efficacy and providing a proof of feasiblity for deployment in practical systems in near future. △ Less

Submitted 2 April, 2024; originally announced April 2024.

Local magnetic response of superconducting Sr$\mathrm{_2}$RuO$\mathrm{_4}$ thin films and rings

Authors: G. M. Ferguson, Hari P. Nair, Nathaniel J. Schreiber, Ludi Miao, Kyle M. Shen, Darrell G. Schlom, Katja C. Nowack

Abstract: We conduct local magnetic measurements on superconducting thin-film samples of Sr$\mathrm{_2}$RuO$\mathrm{_4}$ using scanning Superconducting Quantum Interference Device (SQUID) susceptometry. From the diamagnetic response, we extract the magnetic penetration depth, $λ$, which exhibits a quadratic temperature dependence at low temperatures. Although a quadratic dependence in high-purity bulk sampl… ▽ More We conduct local magnetic measurements on superconducting thin-film samples of Sr$\mathrm{_2}$RuO$\mathrm{_4}$ using scanning Superconducting Quantum Interference Device (SQUID) susceptometry. From the diamagnetic response, we extract the magnetic penetration depth, $λ$, which exhibits a quadratic temperature dependence at low temperatures. Although a quadratic dependence in high-purity bulk samples has been attributed to non-local electrodynamics, our analysis suggests that in our thin-film samples the presence of scattering is the origin of the quadratic dependence. While we observe micron-scale variations in the diamagnetic response and superconducting transition temperature, the form of the temperature dependence of $λ$ is independent of position. Finally, we characterize flux trapping in superconducting rings lithographically fabricated from the thin films, paving the way to systematic device-based tests of the superconducting order parameter in Sr$\mathrm{_2}$RuO$\mathrm{_4}$. △ Less

Submitted 25 March, 2024; originally announced March 2024.

Controllable suppression of the unconventional superconductivity in bulk and thin-film Sr$_{2}$RuO$_{4}$ via high-energy electron irradiation

Authors: Jacob P. Ruf, Hilary M. L. Noad, Romain Grasset, Ludi Miao, Elina Zhakina, Philippa H. McGuinness, Hari P. Nair, Nathaniel J. Schreiber, Naoki Kikugawa, Dmitry Sokolov, Marcin Konczykowski, Darrell G. Schlom, Kyle M. Shen, Andrew P. Mackenzie

Abstract: In bulk Sr$_{2}$RuO$_{4}$, the strong sensitivity of the superconducting transition temperature $T_{\text{c}}$ to nonmagnetic impurities provides robust evidence for a superconducting order parameter that changes sign around the Fermi surface. In superconducting epitaxial thin-film Sr$_{2}$RuO$_{4}$, the relationship between $T_{\text{c}}$ and the residual resistivity $ρ_0$, which in bulk samples… ▽ More In bulk Sr$_{2}$RuO$_{4}$, the strong sensitivity of the superconducting transition temperature $T_{\text{c}}$ to nonmagnetic impurities provides robust evidence for a superconducting order parameter that changes sign around the Fermi surface. In superconducting epitaxial thin-film Sr$_{2}$RuO$_{4}$, the relationship between $T_{\text{c}}$ and the residual resistivity $ρ_0$, which in bulk samples is taken to be a proxy for the low-temperature elastic scattering rate, is far less clear. Using high-energy electron irradiation to controllably introduce point disorder into bulk single-crystal and thin-film Sr$_{2}$RuO$_{4}$, we show that $T_{\text{c}}$ is suppressed in both systems at nearly identical rates. This suggests that part of $ρ_0$ in films comes from defects that do not contribute to superconducting pairbreaking, and establishes a quantitative link between the superconductivity of bulk and thin-film samples. △ Less

Submitted 29 February, 2024; originally announced February 2024.

OzMAC: An Energy-Efficient Sparsity-Exploiting Multiply-Accumulate-Unit Design for DL Inference

Authors: Harideep Nair, Prabhu Vellaisamy, Tsung-Han Lin, Perry Wang, Shawn Blanton, John Paul Shen

Abstract: General Matrix Multiply (GEMM) hardware, employing large arrays of multiply-accumulate (MAC) units, perform bulk of the computation in deep learning (DL). Recent trends have established 8-bit integer (INT8) as the most widely used precision for DL inference. This paper proposes a novel MAC design capable of dynamically exploiting bit sparsity (i.e., number of `0' bits within a binary value) in inp… ▽ More General Matrix Multiply (GEMM) hardware, employing large arrays of multiply-accumulate (MAC) units, perform bulk of the computation in deep learning (DL). Recent trends have established 8-bit integer (INT8) as the most widely used precision for DL inference. This paper proposes a novel MAC design capable of dynamically exploiting bit sparsity (i.e., number of `0' bits within a binary value) in input data to achieve significant improvements on area, power and energy. The proposed architecture, called OzMAC (Omit-zero-MAC), skips over zeros within a binary input value and performs simple shift-and-add-based compute in place of expensive multipliers. We implement OzMAC in SystemVerilog and present post-synthesis performance-power-area (PPA) results using commercial TSMC N5 (5nm) process node. Using eight pretrained INT8 deep neural networks (DNNs) as benchmarks, we demonstrate the existence of high bit sparsity in real DNN workloads and show that 8-bit OzMAC improves all three metrics of area, power, and energy significantly by 21%, 70%, and 28%, respectively. Similar improvements are achieved when scaling data precisions (4, 8, 16 bits) and clock frequencies (0.5 GHz, 1 GHz, 1.5 GHz). For the 8-bit OzMAC, scaling its frequency to normalize the throughput relative to conventional MAC, it still achieves 30% improvement on both power and energy. △ Less

Submitted 29 February, 2024; originally announced February 2024.

Scalable Multi-modal Model Predictive Control via Duality-based Interaction Predictions

Authors: Hansung Kim, Siddharth H. Nair, Francesco Borrelli

Abstract: We propose a hierarchical architecture designed for scalable real-time Model Predictive Control (MPC) in complex, multi-modal traffic scenarios. This architecture comprises two key components: 1) RAID-Net, a novel attention-based Recurrent Neural Network that predicts relevant interactions along the MPC prediction horizon between the autonomous vehicle and the surrounding vehicles using Lagrangian… ▽ More We propose a hierarchical architecture designed for scalable real-time Model Predictive Control (MPC) in complex, multi-modal traffic scenarios. This architecture comprises two key components: 1) RAID-Net, a novel attention-based Recurrent Neural Network that predicts relevant interactions along the MPC prediction horizon between the autonomous vehicle and the surrounding vehicles using Lagrangian duality, and 2) a reduced Stochastic MPC problem that eliminates irrelevant collision avoidance constraints, enhancing computational efficiency. Our approach is demonstrated in a simulated traffic intersection with interactive surrounding vehicles, showcasing a 12x speed-up in solving the motion planning problem. A video demonstrating the proposed architecture in multiple complex traffic scenarios can be found here: https://youtu.be/-pRiOnPb9_c. GitHub: https://github.com/MPC-Berkeley/hmpc_raidnet △ Less

Submitted 2 June, 2024; v1 submitted 1 February, 2024; originally announced February 2024.

Comments: Accepted at IEEE Intelligent Vehicles Symposium 2024

Energy Relaxation and dynamics in the correlated metal Sr$_2$RuO$_4$ via THz two-dimensional coherent spectroscopy

Authors: David Barbalas, Ralph Romero III, Dipanjan Chaudhuri, Fahad Mahmood, Hari P. Nair, Nathaniel J. Schreiber, Darrel G. Schlom, K. M. Shen, N. P. Armitage

Abstract: Separating out the contributions of different scattering channels in strongly interacting metals is crucial in identifying the mechanisms that govern their properties. While momentum or current relaxation rates can be readily probed via \textit{dc} resistivity or optical/THz spectroscopy, distinguishing different kinds of inelastic scattering can be more challenging. Using nonlinear THz 2D coheren… ▽ More Separating out the contributions of different scattering channels in strongly interacting metals is crucial in identifying the mechanisms that govern their properties. While momentum or current relaxation rates can be readily probed via \textit{dc} resistivity or optical/THz spectroscopy, distinguishing different kinds of inelastic scattering can be more challenging. Using nonlinear THz 2D coherent spectroscopy, we measure the rates of energy relaxation after THz excitation in the strongly interacting Fermi liquid, Sr$_2$RuO$_4$. Energy relaxation is a bound on the total scattering and specifically a measure of contributions to the electron self-energy that arise from {\it inelastic} coupling to a bath. We observe two distinct energy relaxation channels: a fast process that we interpret as energy loss to the phonon system and a much slower relaxation that we interpret as arising from a non-equilibrium phonon effects and subsequent heat loss through diffusion. Interestingly, even the faster energy relaxation rate is at least an order of magnitude slower than the overall momentum relaxation rate, consistent with strong electron interactions and the dominance of energy-conserving umklapp or interband electron-electron scattering in momentum relaxation. The slowest energy relaxation rate decays on a sub-GHz scale, consistent with the relaxation dynamics of non-equilibrium phonons. Our observations reveal the versatility of nonlinear THz spectroscopy to measure the energy relaxation dynamics in correlated metals. Our work also highlights the need for improved theoretical understanding of such processes in interacting metals. △ Less

Submitted 20 December, 2023; originally announced December 2023.

Comments: 6 pages, 3 figures. SM included

Accumulation and removal of Si impurities on $β-Ga_2O_3$ arising from ambient air exposure

Authors: J. P. McCandless, C. A. Gorsak, V. Protasenko, D. G. Schlom, Michael O. Thompson, H. G. Xing, D. Jena, H. P. Nair

Abstract: Here we report that the source of Si impurities commonly observed on (010) $β-Ga_2O_3$ is from exposure of the surface to air. Moreover, we find that a 15 minute HF (49%) treatment reduces the Si density by approximately 1 order of magnitude on (010) $β-Ga_2O_3$ surfaces. This reduction in Si is critical for the elimination of the often observed parasitic conducting channel, which negatively affec… ▽ More Here we report that the source of Si impurities commonly observed on (010) $β-Ga_2O_3$ is from exposure of the surface to air. Moreover, we find that a 15 minute HF (49%) treatment reduces the Si density by approximately 1 order of magnitude on (010) $β-Ga_2O_3$ surfaces. This reduction in Si is critical for the elimination of the often observed parasitic conducting channel, which negatively affects transport properties and lateral transistor performance. After the HF treatment the sample must be immediately put under vacuum, for the Si fully returns within 10 minutes of additional air exposure. Lastly, we demonstrate that performing a 30 minute HF (49%) treatment on the substrate before growth has no deleterious effect on the structure or on the epitaxy surface after subsequent $Ga_2O_3$ growth. △ Less

Submitted 11 December, 2023; originally announced December 2023.

Spontaneous supercrystal formation during a strain-engineered metal-insulator transition

Authors: O. Yu. Gorobtsov, L. Miao, Z. Shao, Y. Tan, N. I. Schnitzer, B. H. Goodge, J. Ruf, D. Weinstock, M. Cherukara, M. V. Holt, H. Nair, L. -Q. Chen, L. F. Kourkoutis, D. G. Schlom, K. M. Shen, A. Singer

Abstract: Mott metal-insulator transitions possess electronic, magnetic, and structural degrees of freedom promising next generation energy-efficient electronics. We report a previously unknown, hierarchically ordered state during a Mott transition and demonstrate correlated switching of functional electronic properties. We elucidate in-situ formation of an intrinsic supercrystal in a Ca2RuO4 thin film. Mac… ▽ More Mott metal-insulator transitions possess electronic, magnetic, and structural degrees of freedom promising next generation energy-efficient electronics. We report a previously unknown, hierarchically ordered state during a Mott transition and demonstrate correlated switching of functional electronic properties. We elucidate in-situ formation of an intrinsic supercrystal in a Ca2RuO4 thin film. Machine learning-assisted X-ray nanodiffraction together with electron microscopy reveal multi-scale periodic domain formation at and below the film transition temperature (TFilm ~ 200-250 K) and a separate anisotropic spatial structure at and above TFilm. Local resistivity measurements imply an intrinsic coupling of the supercrystal orientation to the material's anisotropic conductivity. Our findings add an additional degree of complexity to the physical understanding of Mott transitions, opening opportunities for designing materials with tunable electronic properties. △ Less

Submitted 20 November, 2023; originally announced November 2023.

Silicon Implantation and Annealing in $β$-Ga$_2$O$_3$: Role of Ambient, Temperature, and Time

Authors: K. R. Gann, N. Pieczulewski1, C. A. Gorsak, K. Heinselman, T. J. Asel, B. A. Noesges, K. T. Smith, D. M. Dryden, H. G. Xing, H. P. Nair, D. A. Muller, M. O. Thompson

Abstract: Optimizing thermal anneals of Si-implanted $β$-Ga$_2$O$_3$ is critical for low resistance contacts and selective area doping. We report the impact of annealing ambient, temperature, and time on activation of room temperature ion-implanted Si in $β$-Ga$_2$O$_3$ at concentrations from 5x10$^{18}$ to 1x10$^{20}$ cm$^{-3}$, demonstrating full activation (>80% activation, mobilities >70 cm$^{2}$/Vs) wi… ▽ More Optimizing thermal anneals of Si-implanted $β$-Ga$_2$O$_3$ is critical for low resistance contacts and selective area doping. We report the impact of annealing ambient, temperature, and time on activation of room temperature ion-implanted Si in $β$-Ga$_2$O$_3$ at concentrations from 5x10$^{18}$ to 1x10$^{20}$ cm$^{-3}$, demonstrating full activation (>80% activation, mobilities >70 cm$^{2}$/Vs) with contact resistances below 0.29 $Ω$-mm. Homoepitaxial $β$-Ga$_2$O$_3$ films, grown by plasma assisted MBE on Fe-doped (010) substrates, were implanted at multiple energies to yield 100 nm box profiles of 5x10$^{18}$, 5x10$^{19}$, and 1x10$^{20}$ cm$^{-3}$. Anneals were performed in a UHV-compatible quartz furnace at 1 bar with well-controlled gas composition. To maintain $β$-Ga$_2$O$_3$ stability, $p_{O2}$ must be greater than 10$^{-9}$ bar. Anneals up to $p_{O2}$ = 1 bar achieve full activation at 5x10$^{18}$ cm$^{-3}$, while 5x10$^{19}$ cm$^{-3}$ must be annealed with $p_{O2}$ <10$^{-4}$ bar and 1x10$^{20}$ cm$^{-3}$ requires $p_{O2}$ <10$^{-6}$ bar. Water vapor prevents activation and must be maintained below 10$^{-8}$ bar. Activation is achieved for anneal temperatures as low as 850 °C with mobility increasing with anneal temperature up to 1050 °C, though Si diffusion has been reported above 950 °C. At 950 °C, activation is maximized between 5 and 20 minutes with longer times resulting in decreased carrier activation (over-annealing). This over-annealing is significant for concentrations above 5x10$^{19}$ cm$^{-3}$ and occurs rapidly at 1x10$^{20}$ cm$^{-3}$. RBS (channeling) suggests damage recovery is seeded from remnant aligned $β$-Ga$_2$O$_3$ that remains after implantation; this conclusion is also supported by STEM showing retention of the $β$-phase with inclusions that resemble the $γ$-phase. △ Less

Submitted 1 November, 2023; originally announced November 2023.

Predictive Control for Autonomous Driving with Uncertain, Multi-modal Predictions

Authors: Siddharth H. Nair, Hotae Lee, Eunhyek Joa, Yan Wang, H. Eric Tseng, Francesco Borrelli

Abstract: We propose a Stochastic MPC (SMPC) formulation for path planning with autonomous vehicles in scenarios involving multiple agents with multi-modal predictions. The multi-modal predictions capture the uncertainty of urban driving in distinct modes/maneuvers (e.g., yield, keep speed) and driving trajectories (e.g., speed, turning radius), which are incorporated for multi-modal collision avoidance cha… ▽ More We propose a Stochastic MPC (SMPC) formulation for path planning with autonomous vehicles in scenarios involving multiple agents with multi-modal predictions. The multi-modal predictions capture the uncertainty of urban driving in distinct modes/maneuvers (e.g., yield, keep speed) and driving trajectories (e.g., speed, turning radius), which are incorporated for multi-modal collision avoidance chance constraints for path planning. In the presence of multi-modal uncertainties, it is challenging to reliably compute feasible path planning solutions at real-time frequencies ($\geq$ 10 Hz). Our main technological contribution is a convex SMPC formulation that simultaneously (1) optimizes over parameterized feedback policies and (2) allocates risk levels for each mode of the prediction. The use of feedback policies and risk allocation enhances the feasibility and performance of the SMPC formulation against multi-modal predictions with large uncertainty. We evaluate our approach via simulations and road experiments with a full-scale vehicle interacting in closed-loop with virtual vehicles. We consider distinct, multi-modal driving scenarios: 1) Negotiating a traffic light and a fast, tailgating agent, 2) Executing an unprotected left turn at a traffic intersection, and 3) Changing lanes in the presence of multiple agents. For all of these scenarios, our approach reliably computes multi-modal solutions to the path-planning problem at real-time frequencies. △ Less

Submitted 31 October, 2023; originally announced October 2023.

Comments: The first three authors contributed equally

Optimizing Hierarchical Queries for the Attribution Reporting API

Authors: Matthew Dawson, Badih Ghazi, Pritish Kamath, Kapil Kumar, Ravi Kumar, Bo Luan, Pasin Manurangsi, Nishanth Mundru, Harikesh Nair, Adam Sealfon, Shengyu Zhu

Abstract: We study the task of performing hierarchical queries based on summary reports from the {\em Attribution Reporting API} for ad conversion measurement. We demonstrate that methods from optimization and differential privacy can help cope with the noise introduced by privacy guardrails in the API. In particular, we present algorithms for (i) denoising the API outputs and ensuring consistency across di… ▽ More We study the task of performing hierarchical queries based on summary reports from the {\em Attribution Reporting API} for ad conversion measurement. We demonstrate that methods from optimization and differential privacy can help cope with the noise introduced by privacy guardrails in the API. In particular, we present algorithms for (i) denoising the API outputs and ensuring consistency across different levels of the tree, and (ii) optimizing the privacy budget across different levels of the tree. We provide an experimental evaluation of the proposed algorithms on public datasets. △ Less

Submitted 27 November, 2023; v1 submitted 25 August, 2023; originally announced August 2023.

Comments: Appeared at AdKDD 2023 workshop; Final proceedings version

Electronic nematic order in the normal state of strontium ruthenate

Authors: Ryan Russell, Hari P. Nair, Kyle M. Shen, Darrell G. Schlom, John W. Harter

Abstract: Despite significant achievements in characterizing the properties of Sr$_2$RuO$_4$ over the last three decades, the precise nature of its electronic ground state is still unresolved. In this work, we provide a missing piece of the puzzle by uncovering evidence of electronic nematic order in the normal state of Sr$_2$RuO$_4$, revealed by ultrafast time-resolved optical dichroism measurements of uni… ▽ More Despite significant achievements in characterizing the properties of Sr$_2$RuO$_4$ over the last three decades, the precise nature of its electronic ground state is still unresolved. In this work, we provide a missing piece of the puzzle by uncovering evidence of electronic nematic order in the normal state of Sr$_2$RuO$_4$, revealed by ultrafast time-resolved optical dichroism measurements of uniaxially strained thin films. This nematic order, whose domains are aligned by the strain, spontaneously breaks the four-fold rotational symmetry of the crystal. The temperature dependence of the dichroism resembles an Ising-like order parameter, and optical pumping induces a coherent oscillation of its amplitude mode. A microscopic model of intra-unit-cell nematic order is presented, highlighting the importance of Coulomb repulsion between neighboring oxygen $p$-orbitals. The existence of electronic nematic order in the normal state of Sr$_2$RuO$_4$ may have consequences for the form and mechanism of superconductivity in this material. △ Less

Submitted 5 April, 2023; originally announced April 2023.

Journal ref: Physical Review B 108, L081105 (2023)

Picosecond volume expansion drives a later-time insulator-metal transition in a nano-textured Mott Insulator

Authors: Anita Verma, Denis Golež, Oleg Yu. Gorobtsov, Kelson Kaj, Ryan Russell, Jeffrey Z. Kaaret, Erik Lamb, Guru Khalsa, Hari P Nair, Yifei Sun, Ryan Bouck, Nathaniel Schreiber, Jacob P. Ruf, Varun Ramaprasad, Yuya Kubota, Tadashi Togashi, Vladimir A. Stoica, Hari Padmanabhan, John W. Freeland, Nicole A. Benedek, Oleg Shpyrko, John W. Harter, Richard D. Averitt, Darrell G. Schlom, Kyle M. Shen , et al. (2 additional authors not shown)

Abstract: Technology moves towards ever faster switching between different electronic and magnetic states of matter. Manipulating properties at terahertz rates requires accessing the intrinsic timescales of electrons (femtoseconds) and associated phonons (10s of femtoseconds to few picoseconds), which is possible with short-pulse photoexcitation. Yet, in many Mott insulators, the electronic transition is ac… ▽ More Technology moves towards ever faster switching between different electronic and magnetic states of matter. Manipulating properties at terahertz rates requires accessing the intrinsic timescales of electrons (femtoseconds) and associated phonons (10s of femtoseconds to few picoseconds), which is possible with short-pulse photoexcitation. Yet, in many Mott insulators, the electronic transition is accompanied by the nucleation and growth of percolating domains of the changed lattice structure, leading to empirical time scales dominated by slow coarsening dynamics. Here, we use time-resolved X-ray diffraction and reflectivity measurements to investigate the photoinduced insulator-to-metal transition in an epitaxially strained thin film Mott insulator Ca2RuO4. The dynamical transition occurs without observable domain formation and coarsening effects, allowing the study of the intrinsic electronic and lattice dynamics. Above a fluence threshold, the initial electronic excitation drives a fast lattice rearrangement, followed by a slower electronic evolution into a metastable non-equilibrium state. Microscopic calculations based on time-dependent dynamical mean-field theory and semiclassical lattice dynamics within a recently published equilibrium energy landscape picture explain the threshold-behavior and elucidate the delayed onset of the electronic phase transition in terms of kinematic constraints on recombination. Analysis of satellite scattering peaks indicates the persistence of a strain-induced nano-texture in the photoexcited film. This work highlights the importance of combined electronic and structural studies to unravel the physics of dynamic transitions and elucidates the role of strain in tuning the timescales of photoinduced processes. △ Less

Submitted 6 April, 2023; v1 submitted 4 April, 2023; originally announced April 2023.

Learning for Online Mixed-Integer Model Predictive Control with Parametric Optimality Certificates

Authors: Luigi Russo, Siddharth H. Nair, Luigi Glielmo, Francesco Borrelli

Abstract: We propose a supervised learning framework for computing solutions of multi-parametric Mixed Integer Linear Programs (MILPs) that arise in Model Predictive Control. Our approach also quantifies sub-optimality for the computed solutions. Inspired by Branch-and-Bound techniques, the key idea is to train a Neural Network/Random Forest, which for a given parameter, predicts a strategy consisting of (1… ▽ More We propose a supervised learning framework for computing solutions of multi-parametric Mixed Integer Linear Programs (MILPs) that arise in Model Predictive Control. Our approach also quantifies sub-optimality for the computed solutions. Inspired by Branch-and-Bound techniques, the key idea is to train a Neural Network/Random Forest, which for a given parameter, predicts a strategy consisting of (1) a set of Linear Programs (LPs) such that their feasible sets form a partition of the feasible set of the MILP and (2) a candidate integer solution. For control computation and sub-optimality quantification, we solve a set of LPs online in parallel. We demonstrate our approach for a motion planning example and compare against various commercial and open-source mixed-integer programming solvers. △ Less

Submitted 22 March, 2023; v1 submitted 21 March, 2023; originally announced March 2023.

Comments: First two authors contributed equally

Robust Output-Lifted Learning Model Predictive Control

Authors: Siddharth H. Nair, Francesco Borrelli

Abstract: We propose an iterative approach for designing Robust Learning Model Predictive Control (LMPC) policies for a class of nonlinear systems with additive, unmodelled dynamics. The nominal dynamics are assumed to be difference flat, i.e., the state and input can be reconstructed using flat output sequences. For the considered class of systems, we synthesize Robust MPC policies and show how to use hist… ▽ More We propose an iterative approach for designing Robust Learning Model Predictive Control (LMPC) policies for a class of nonlinear systems with additive, unmodelled dynamics. The nominal dynamics are assumed to be difference flat, i.e., the state and input can be reconstructed using flat output sequences. For the considered class of systems, we synthesize Robust MPC policies and show how to use historical trajectory data collected during iterative tasks to 1) obtain bounds on the unmodelled dynamics and 2) construct a convex value function approximation along with a convex safe set in the space of output sequences for designing terminal components in the Robust MPC design. We show that the proposed strategy guarantees robust constraint satisfaction, asymptotic convergence to a desired subset of the state space, and non-decreasing closed-loop performance at each policy update. Finally, simulation results demonstrate the effectiveness of the proposed strategy on a minimum time control problem using a constrained nonlinear and uncertain vehicle model. △ Less

Submitted 21 March, 2023; originally announced March 2023.

Sawtooth lattice multiferroic BeCr$_2$O$_4$: Non-collinear magnetic structure and multiple magnetic transitions

Authors: Hector Cein Mandujano, Alejandro Metta, Neven Barisic, Qiang Zhang, Wojciech Tabis, Naveen Kumar Chogondahalli Muniraju, Harikrishnan S. Nair

Abstract: Noncollinear magnetic structures and multiple magnetic phase transitions in a sawtooth lattice antiferromagnet consisting of Cr$^{3+}$ are experimentally identified in this work, thereby proposing the scenario of magnetism-driven ferroelectricity in a sawtooth lattice. The title compound, BeCr$_2$O$_4$, displays three magnetic phase transitions at low temperatures, at $T_{N1}\approx$ 7.5 K, at… ▽ More Noncollinear magnetic structures and multiple magnetic phase transitions in a sawtooth lattice antiferromagnet consisting of Cr$^{3+}$ are experimentally identified in this work, thereby proposing the scenario of magnetism-driven ferroelectricity in a sawtooth lattice. The title compound, BeCr$_2$O$_4$, displays three magnetic phase transitions at low temperatures, at $T_{N1}\approx$ 7.5 K, at $T_{N2}\approx$ 25 K and at $T_{N3}\approx$ 26 K, revealed through magnetic susceptibility, specific heat and neutron diffraction in this work. These magnetic phase transitions are found to be influenced by externally applied magnetic fields. Isothermal magnetization curves at low temperatures below the magnetic transitions indicate the antiferromagnetic nature of \bco\ with two spin-flop-like transitions occurring at $H_{c1}\approx$ 29 kOe and $H_{c2} \approx$ 47 kOe. Our high-resolution X-ray and neutron diffraction studies, performed on single crystal and powder samples unambiguously determined the crystal structure as orthorhombic $Pbnm$. By performing the magnetic superspace group analysis of the neutron diffraction data at low temperatures, the magnetic structure in the temperature range $T_{N3,N2} < T < T_{N1}$ is determined to be the polar magnetic space group, $P21nm.1^{\prime}(00g)0s0s$ with a cycloidal magnetic propagation vector $\textbf{k}_1$ = (0, 0, 0.090(1)). The magnetic structure in the newly identified phase below $T_{N1}$, is determined as $P21/b.1^{\prime}[b](00g)00s$ with the magnetic propagation vector $\textbf{k}_2$ = (0, 0, 0.908(1)). The cycloidal spin structure determined in our work is usually associated with electric polarization, thereby making \bco\ a promising multiferroic belonging to the sparsely populated family of sawtooth lattice antiferromagnets. △ Less

Submitted 11 July, 2023; v1 submitted 5 March, 2023; originally announced March 2023.

Comments: 24 pages, 5 figures, accepted

Successful Kinetic Impact into an Asteroid for Planetary Defense

Authors: R. Terik Daly, Carolyn M. Ernst, Olivier S. Barnouin, Nancy L. Chabot, Andrew S. Rivkin, Andrew F. Cheng, Elena Y. Adams, Harrison F. Agrusa, Elisabeth D. Abel, Amy L. Alford, Erik I. Asphaug, Justin A. Atchison, Andrew R. Badger, Paul Baki, Ronald-L. Ballouz, Dmitriy L. Bekker, Julie Bellerose, Shyam Bhaskaran, Bonnie J. Buratti, Saverio Cambioni, Michelle H. Chen, Steven R. Chesley, George Chiu, Gareth S. Collins, Matthew W. Cox , et al. (76 additional authors not shown)

Abstract: While no known asteroid poses a threat to Earth for at least the next century, the catalog of near-Earth asteroids is incomplete for objects whose impacts would produce regional devastation. Several approaches have been proposed to potentially prevent an asteroid impact with Earth by deflecting or disrupting an asteroid. A test of kinetic impact technology was identified as the highest priority sp… ▽ More While no known asteroid poses a threat to Earth for at least the next century, the catalog of near-Earth asteroids is incomplete for objects whose impacts would produce regional devastation. Several approaches have been proposed to potentially prevent an asteroid impact with Earth by deflecting or disrupting an asteroid. A test of kinetic impact technology was identified as the highest priority space mission related to asteroid mitigation. NASA's Double Asteroid Redirection Test (DART) mission is the first full-scale test of kinetic impact technology. The mission's target asteroid was Dimorphos, the secondary member of the S-type binary near-Earth asteroid (65803) Didymos. This binary asteroid system was chosen to enable ground-based telescopes to quantify the asteroid deflection caused by DART's impact. While past missions have utilized impactors to investigate the properties of small bodies those earlier missions were not intended to deflect their targets and did not achieve measurable deflections. Here we report the DART spacecraft's autonomous kinetic impact into Dimorphos and reconstruct the impact event, including the timeline leading to impact, the location and nature of the DART impact site, and the size and shape of Dimorphos. The successful impact of the DART spacecraft with Dimorphos and the resulting change in Dimorphos's orbit demonstrates that kinetic impactor technology is a viable technique to potentially defend Earth if necessary. △ Less

Submitted 3 March, 2023; originally announced March 2023.

Comments: Accepted by Nature

Interaction and Decision Making-aware Motion Planning using Branch Model Predictive Control

Authors: Rui Oliveira, Siddharth H. Nair, Bo Wahlberg

Abstract: Motion planning for autonomous vehicles sharing the road with human drivers remains challenging. The difficulty arises from three challenging aspects: human drivers are 1) multi-modal, 2) interacting with the autonomous vehicle, and 3) actively making decisions based on the current state of the traffic scene. We propose a motion planning framework based on Branch Model Predictive Control to deal w… ▽ More Motion planning for autonomous vehicles sharing the road with human drivers remains challenging. The difficulty arises from three challenging aspects: human drivers are 1) multi-modal, 2) interacting with the autonomous vehicle, and 3) actively making decisions based on the current state of the traffic scene. We propose a motion planning framework based on Branch Model Predictive Control to deal with these challenges. The multi-modality is addressed by considering multiple future outcomes associated with different decisions taken by the human driver. The interactive nature of humans is considered by modeling them as reactive agents impacted by the actions of the autonomous vehicle. Finally, we consider a model developed in human neuroscience studies as a possible way of encoding the decision making process of human drivers. We present simulation results in various scenarios, showing the advantages of the proposed method and its ability to plan assertive maneuvers that convey intent to humans. △ Less

Submitted 31 January, 2023; originally announced February 2023.

Comments: 8 pages, 10 figures

Silicon-doped $β$-Ga$_2$O$_3$ films grown at 1 $μ$m/h by suboxide molecular-beam epitaxy

Authors: Kathy Azizie, Felix V. E. Hensling, Cameron A. Gorsak, Yunjo Kim, Daniel M. Dryden, M. K. Indika Senevirathna, Selena Coye, Shun-Li Shang, Jacob Steele, Patrick Vogt, Nicholas A. Parker, Yorick A. Birkhölzer, Jonathan P. McCandless, Debdeep Jena, Huili G. Xing, Zi-Kui Liu, Michael D. Williams, Andrew J. Green, Kelson Chabak, Adam T. Neal, Shin Mou, Michael O. Thompson, Hari P. Nair, Darrell G. Schlom

Abstract: We report the use of suboxide molecular-beam epitaxy (S-MBE) to grow $β$-Ga$_2$O$_3$ at a growth rate of ~1 $μ$m/h with control of the silicon doping concentration from 5x10$^{16}$ to 10$^{19}$ cm$^{-3}$. In S-MBE, pre-oxidized gallium in the form of a molecular beam that is 99.98\% Ga$_2$O, i.e., gallium suboxide, is supplied. Directly supplying Ga2O to the growth surface bypasses the rate-limiti… ▽ More We report the use of suboxide molecular-beam epitaxy (S-MBE) to grow $β$-Ga$_2$O$_3$ at a growth rate of ~1 $μ$m/h with control of the silicon doping concentration from 5x10$^{16}$ to 10$^{19}$ cm$^{-3}$. In S-MBE, pre-oxidized gallium in the form of a molecular beam that is 99.98\% Ga$_2$O, i.e., gallium suboxide, is supplied. Directly supplying Ga2O to the growth surface bypasses the rate-limiting first step of the two-step reaction mechanism involved in the growth of $β$-Ga$_2$O$_3$ by conventional MBE. As a result, a growth rate of ~1 $μ$m/h is readily achieved at a relatively low growth temperature (T$_{sub}$ = 525 $^\circ$C), resulting in films with high structural perfection and smooth surfaces (rms roughness of < 2 nm on ~1 $μ$m thick films). Silicon-containing oxide sources (SiO and SiO$_2$) producing an SiO suboxide molecular beam are used to dope the $β$-Ga$_2$O$_3$ layers. Temperature-dependent Hall effect measurements on a 1 $μ$m thick film with a mobile carrier concentration of 2.7x10$^{17}$ cm$^{-3}$ reveal a room-temperature mobility of 124 cm$^2$ V$^{-1}$ s$^{-1}$ that increases to 627 cm$^2$ V$^{-1}$ s$^{-1}$ at 76 K; the silicon dopants are found to exhibit an activation energy of 27 meV. We also demonstrate working MESFETs made from these silicon-doped $β$-Ga$_2$O$_3$ films grown by S-MBE at growth rates of ~1 $μ$m/h. △ Less

Submitted 22 December, 2022; originally announced December 2022.

Comments: 19 pages, 7 figures, 2 tables, 2 pages supplementary materials

Real-space imaging of polar and elastic nano-textures in thin films via inversion of diffraction data

Authors: Ziming Shao, Noah Schnitzer, Jacob Ruf, Oleg Y. Gorobtsov, Cheng Dai, Berit H. Goodge, Tiannan Yang, Hari Nair, Vlad A. Stoica, John W. Freeland, Jacob Ruff, Long-Qing Chen, Darrell G. Schlom, Kyle M. Shen, Lena F. Kourkoutis, Andrej Singer

Abstract: Exploiting the emerging nanoscale periodicities in epitaxial, single-crystal thin films is an exciting direction in quantum materials science: confinement and periodic distortions induce novel properties. The structural motifs of interest are ferroelastic, ferroelectric, multiferroic, and, more recently, topologically protected magnetization and polarization textures. A critical step towards heter… ▽ More Exploiting the emerging nanoscale periodicities in epitaxial, single-crystal thin films is an exciting direction in quantum materials science: confinement and periodic distortions induce novel properties. The structural motifs of interest are ferroelastic, ferroelectric, multiferroic, and, more recently, topologically protected magnetization and polarization textures. A critical step towards heterostructure engineering is understanding their nanoscale structure, best achieved through real-space imaging. X-ray Bragg coherent diffractive imaging visualizes sub-picometer crystalline displacements with tens of nanometers spatial resolution. Yet, it is limited to objects spatially confined in all three dimensions and requires highly coherent, laser-like x-rays. Here we lift the confinement restriction by developing real-space imaging of periodic lattice distortions: we combine an iterative phase retrieval algorithm with unsupervised machine learning to invert the diffuse scattering in conventional x-ray reciprocal-space mapping into real-space images of polar and elastic textures in thin epitaxial films. We first demonstrate our imaging in PbTiO3/SrTiO3 superlattices to be consistent with published phase-field model calculations. We then visualize strain-induced ferroelastic domains emerging during the metal-insulator transition in Ca2RuO4 thin films. Instead of homogeneously transforming into a low-temperature structure (like in bulk), the strained Mott insulator splits into nanodomains with alternating lattice constants, as confirmed by cryogenic scanning transmission electron microscopy. Our study reveals the type, size, orientation, and crystal displacement field of the nano-textures. The non-destructive imaging of textures promises to improve models for their dynamics and enable advances in quantum materials and microelectronics. △ Less

Submitted 2 November, 2022; originally announced November 2022.

Control of Uncertain PWA Systems using Difference-of-Convex Decompositions

Authors: Siddharth H. Nair, Yvonne R. Stürz

Abstract: In this report, we analyze and design feedback policies for discrete-time Piecewise-Affine (PWA) systems with uncertainty in both the affine dynamics and the polytopic partition. The main idea is to utilise the Difference-of-Convex (DC) decomposition of continuous PWA systems to derive quadratic Lyapunov functions as stability certificates and stabilizing affine policies in a higher dimensional sp… ▽ More In this report, we analyze and design feedback policies for discrete-time Piecewise-Affine (PWA) systems with uncertainty in both the affine dynamics and the polytopic partition. The main idea is to utilise the Difference-of-Convex (DC) decomposition of continuous PWA systems to derive quadratic Lyapunov functions as stability certificates and stabilizing affine policies in a higher dimensional space. When projected back to the state space, we obtain time-varying PWQ Lyapunov functions and time-varying PWA feedback policies. △ Less

Submitted 20 March, 2023; v1 submitted 26 September, 2022; originally announced September 2022.

Comments: Extended project report

Advertising Media and Target Audience Optimization via High-dimensional Bandits

Authors: Wenjia Ba, J. Michael Harrison, Harikesh S. Nair

Abstract: We present a data-driven algorithm that advertisers can use to automate their digital ad-campaigns at online publishers. The algorithm enables the advertiser to search across available target audiences and ad-media to find the best possible combination for its campaign via online experimentation. The problem of finding the best audience-ad combination is complicated by a number of distinctive chal… ▽ More We present a data-driven algorithm that advertisers can use to automate their digital ad-campaigns at online publishers. The algorithm enables the advertiser to search across available target audiences and ad-media to find the best possible combination for its campaign via online experimentation. The problem of finding the best audience-ad combination is complicated by a number of distinctive challenges, including (a) a need for active exploration to resolve prior uncertainty and to speed the search for profitable combinations, (b) many combinations to choose from, giving rise to high-dimensional search formulations, and (c) very low success probabilities, typically just a fraction of one percent. Our algorithm (designated LRDL, an acronym for Logistic Regression with Debiased Lasso) addresses these challenges by combining four elements: a multiarmed bandit framework for active exploration; a Lasso penalty function to handle high dimensionality; an inbuilt debiasing kernel that handles the regularization bias induced by the Lasso; and a semi-parametric regression model for outcomes that promotes cross-learning across arms. The algorithm is implemented as a Thompson Sampler, and to the best of our knowledge, it is the first that can practically address all of the challenges above. Simulations with real and synthetic data show the method is effective and document its superior performance against several benchmarks from the recent high-dimensional bandit literature. △ Less

Submitted 17 September, 2022; originally announced September 2022.

Comments: 39 pages, 8 figures

Using blogs to make peer-reviewed research more accessible

Authors: Nicholas T. Young, Briley L. Lewis, Emily Kerr, Prasanth H. Nair

Abstract: Discipline-based education researchers produce knowledge that aims to help instructors improve student learning and educational outcomes. Yet, the information produced may not even reach the educators it is intended to influence. Prior work has found that instructors often face barriers to implementing practices in peer-reviewed literature. Some of these barriers are related to accessing the knowl… ▽ More Discipline-based education researchers produce knowledge that aims to help instructors improve student learning and educational outcomes. Yet, the information produced may not even reach the educators it is intended to influence. Prior work has found that instructors often face barriers to implementing practices in peer-reviewed literature. Some of these barriers are related to accessing the knowledge in the first place such as difficulty finding and understanding research and a lack of time to do so. To lower these barriers, we created an online blog, PERbites, that summarizes recent discipline-based education research in short posts that use plain language. Having covered nearly 100 papers to date, we conducted a survey to see if we were addressing the need we had originally set out to address. We posted a 23-item survey on our website and received 24 usable responses. The results suggested that readers do generally agree that we are meeting our original goals. Readers reported that our articles were easier to understand and used more plain language than a typical discipline-based education research (DBER) journal article. At the same time, readers thought that all the important information was still included. Finally, readers said that this approach helped them keep up with DBER studies and read about papers they otherwise would not have. However, most readers did not indicate they changed their teaching and research practice as a result of reading our blog. Our results suggest that alternative methods of sharing research (e.g., non-peer reviewed publications or conference talks) can be an effective method of connecting research with practitioners, and future work should consider how we as a community might build on these efforts to ensure education research can make meaningful changes in the classroom. △ Less

Submitted 15 September, 2022; originally announced September 2022.

Comments: Published in the Proceedings of the 2022 Physics Education Research Conference, Grand Rapids, MI, US July 13th - July 14th

Atomic-scale mapping and quantification of local Ruddlesden-Popper phase variations

Authors: Erin E. Fleck, Berit H. Goodge, Matthew R. Barone, Hari P. Nair, Nathaniel J. Schreiber, Natalie M. Dawley, Darrell G. Schlom, Lena F. Kourkoutis

Abstract: The Ruddlesden-Popper ($A_{n+1}B_{n}\text{O}_{3n+1}$) compounds are a highly tunable class of materials whose functional properties can be dramatically impacted by their structural phase $n$. The negligible energetic differences associated with forming a sample with a single value of $n$ versus a mixture of $n$ makes the growth of these materials difficult to control and can lead to local atomic-s… ▽ More The Ruddlesden-Popper ($A_{n+1}B_{n}\text{O}_{3n+1}$) compounds are a highly tunable class of materials whose functional properties can be dramatically impacted by their structural phase $n$. The negligible energetic differences associated with forming a sample with a single value of $n$ versus a mixture of $n$ makes the growth of these materials difficult to control and can lead to local atomic-scale structural variation arising from small stoichiometric deviations. In this work, we present a Python analysis platform to detect, measure, and quantify the presence of different $n$-phases based on atomic-resolution scanning transmission electron microscopy (STEM) images in a statistically rigorous manner. We employ phase analysis on the 002 Bragg peak to identify horizontal Ruddlesden-Popper faults which appear as regions of high positive compressive strain within the lattice image, allowing us to quantify the local structure. Our semi-automated technique offers statistical advantages by considering effects of finite projection thickness, limited fields of view, and precise sampling rates. This method retains the real-space distribution of layer variations allowing for a spatial mapping of local $n$-phases, enabling both quantification of intergrowth occurrence as well as qualitative description of their distribution, opening the door to new insights and levels of control over a range of layered materials. △ Less

Submitted 18 August, 2022; originally announced August 2022.

Comments: 8 pages, 5 figures

Collision Avoidance for Dynamic Obstacles with Uncertain Predictions using Model Predictive Control

Authors: Siddharth H. Nair, Eric H. Tseng, Francesco Borrelli

Abstract: We propose a Model Predictive Control (MPC) for collision avoidance between an autonomous agent and dynamic obstacles with uncertain predictions. The collision avoidance constraints are imposed by enforcing positive distance between convex sets representing the agent and the obstacles, and tractably reformulating them using Lagrange duality. This approach allows for smooth collision avoidance cons… ▽ More We propose a Model Predictive Control (MPC) for collision avoidance between an autonomous agent and dynamic obstacles with uncertain predictions. The collision avoidance constraints are imposed by enforcing positive distance between convex sets representing the agent and the obstacles, and tractably reformulating them using Lagrange duality. This approach allows for smooth collision avoidance constraints even for polytopes, which otherwise require mixed-integer or non-smooth constraints. We consider three widely used descriptions of the uncertain obstacle position: 1) Arbitrary distribution with polytopic support, 2) Gaussian distributions and 3) Arbitrary distribution with first two moments known. For each case we obtain deterministic reformulations of the collision avoidance constraints. The proposed MPC formulation optimizes over feedback policies to reduce conservatism in satisfying the collision avoidance constraints. The proposed approach is validated using simulations of traffic intersections in CARLA. △ Less

Submitted 6 August, 2022; originally announced August 2022.

Comments: Accepted to CDC'22

Stochastic MPC with Dual Control for Autonomous Driving with Multi-Modal Interaction-Aware Predictions

Authors: Siddharth H. Nair, Vijay Govindarajan, Theresa Lin, Yan Wang, Eric H. Tseng, Francesco Borrelli

Abstract: We propose a Stochastic MPC (SMPC) approach for autonomous driving which incorporates multi-modal, interaction-aware predictions of surrounding vehicles. For each mode, vehicle motion predictions are obtained by a control model described using a basis of fixed features with unknown weights. The proposed SMPC formulation finds optimal controls which serves two purposes: 1) reducing conservatism of… ▽ More We propose a Stochastic MPC (SMPC) approach for autonomous driving which incorporates multi-modal, interaction-aware predictions of surrounding vehicles. For each mode, vehicle motion predictions are obtained by a control model described using a basis of fixed features with unknown weights. The proposed SMPC formulation finds optimal controls which serves two purposes: 1) reducing conservatism of the SMPC by optimizing over parameterized control laws and 2) prediction and estimation of feature weights used in interaction-aware modeling using Kalman filtering. The proposed approach is demonstrated on a longitudinal control example, with uncertainties in predictions of the autonomous and surrounding vehicles. △ Less

Submitted 6 August, 2022; originally announced August 2022.

Comments: Accepted to AVEC'22

TNN7: A Custom Macro Suite for Implementing Highly Optimized Designs of Neuromorphic TNNs

Authors: Harideep Nair, Prabhu Vellaisamy, Santha Bhasuthkar, John Paul Shen

Abstract: Temporal Neural Networks (TNNs), inspired from the mammalian neocortex, exhibit energy-efficient online sensory processing capabilities. Recent works have proposed a microarchitecture framework for implementing TNNs and demonstrated competitive performance on vision and time-series applications. Building on these previous works, this work proposes TNN7, a suite of nine highly optimized custom macr… ▽ More Temporal Neural Networks (TNNs), inspired from the mammalian neocortex, exhibit energy-efficient online sensory processing capabilities. Recent works have proposed a microarchitecture framework for implementing TNNs and demonstrated competitive performance on vision and time-series applications. Building on these previous works, this work proposes TNN7, a suite of nine highly optimized custom macros developed using a predictive 7nm Process Design Kit (PDK), to enhance the efficiency, modularity and flexibility of the TNN design framework. TNN prototypes for two applications are used for evaluation of TNN7. An unsupervised time-series clustering TNN delivering competitive performance can be implemented within 40 uW power and 0.05 mm^2 area, while a 4-layer TNN that achieves an MNIST error rate of 1% consumes only 18 mW and 24.63 mm^2. On average, the proposed macros reduce power, delay, area, and energy-delay product by 14%, 16%, 28%, and 45%, respectively. Furthermore, employing TNN7 significantly reduces the synthesis runtime of TNN designs (by more than 3x), allowing for highly-scaled TNN implementations to be realized. △ Less

Submitted 25 May, 2022; v1 submitted 15 May, 2022; originally announced May 2022.

Comments: To be published in ISVLSI 2022

Theory, phenomenology, and experimental avenues for dark showers: a Snowmass 2021 report

Authors: Guillaume Albouy, Jared Barron, Hugues Beauchesne, Elias Bernreuther, Marcella Bona, Cesare Cazzaniga, Cari Cesarotti, Timothy Cohen, Annapaola de Cosa, David Curtin, Zeynep Demiragli, Caterina Doglioni, Alison Elliot, Karri Folan DiPetrillo, Florian Eble, Carlos Erice, Chad Freer, Aran Garcia-Bellido, Caleb Gemmell, Marie-Hélène Genest, Giovanni Grilli di Cortona, Giuliano Gustavino, Nicoline Hemme, Tova Holmes, Deepak Kar , et al. (29 additional authors not shown)

Abstract: In this work, we consider the case of a strongly coupled dark/hidden sector, which extends the Standard Model (SM) by adding an additional non-Abelian gauge group. These extensions generally contain matter fields, much like the SM quarks, and gauge fields similar to the SM gluons. We focus on the exploration of such sectors where the dark particles are produced at the LHC through a portal and unde… ▽ More In this work, we consider the case of a strongly coupled dark/hidden sector, which extends the Standard Model (SM) by adding an additional non-Abelian gauge group. These extensions generally contain matter fields, much like the SM quarks, and gauge fields similar to the SM gluons. We focus on the exploration of such sectors where the dark particles are produced at the LHC through a portal and undergo rapid hadronization within the dark sector before decaying back, at least in part and potentially with sizeable lifetimes, to SM particles, giving a range of possibly spectacular signatures such as emerging or semi-visible jets. Other, non-QCD-like scenarios leading to soft unclustered energy patterns or glueballs are also discussed. After a review of the theory, existing benchmarks and constraints, this work addresses how to build consistent benchmarks from the underlying physical parameters and present new developments for the PYTHIA Hidden Valley module, along with jet substructure studies. Finally, a series of improved search strategies is presented in order to pave the way for a better exploration of the dark showers at the LHC. △ Less

Submitted 27 June, 2022; v1 submitted 17 March, 2022; originally announced March 2022.

Comments: Uniform notation, fixed typos, improved numerical analysis, added references, comments welcome

A ferrotoroidic candidate with well-separated spin chains

Authors: Jun Zhang, Xiancheng Wang, Long Zhou, Gangxiu Liu, Devashibhai T. Adroja, Ivan da Silva, Franz Demmel, Dmitry Khalyavin, Jhuma Sannigrahi, Hari S. Nair, Lei Duan, Jianfa Zhao, Zeng Deng, Runze Yu, Xi Shen, Richeng Yu, Hui Zhao, Jimin Zhao, Youwen Long, Zhiwei Hu, Hong-Ji Lin, Ting-Shan Chan, Chien-Te Chen, Wei Wu, Changqing Jin

Abstract: The search of novel quasi one-dimensional (1D) materials is one of the important aspects in the field of material science. Toroidal moment, the order parameter of ferrotoroidic order, can be generated by a head-to-tail configuration of magnetic moment. It has been theoretically proposed that one-dimensional (1D) dimerized and antiferromagnetic-like spin chain hosts ferrotoroidicity and has the tor… ▽ More The search of novel quasi one-dimensional (1D) materials is one of the important aspects in the field of material science. Toroidal moment, the order parameter of ferrotoroidic order, can be generated by a head-to-tail configuration of magnetic moment. It has been theoretically proposed that one-dimensional (1D) dimerized and antiferromagnetic-like spin chain hosts ferrotoroidicity and has the toroidal moment composed of only two antiparallel spins. Here, we report a ferrotoroidic candidate of Ba6Cr2S10 with such a theoretical model of spin chain. The structure consists of unique dimerized face-sharing CrS6 octahedral chains along the c axis. An antiferromagnetic-like ordering at ~10 K breaks both space- and time-reversal symmetries and the magnetic point group of mm'2' allows three ferroic orders in Ba6Cr2S10: (anti)ferromagnetic, ferroelectric and ferrotoroidic orders. Our investigation reveals that Ba6Cr2S10 is a rare ferrotoroidic candidate with quasi 1D spin chain, which can be considered as a starting point for the further exploration of the physics and applications of ferrotoroidicity. △ Less

Submitted 21 January, 2022; originally announced January 2022.

Strain-induced orbital energy shift in antiferromagnetic RuO2 revealed by resonant elastic x-ray scattering

Authors: Benjamin Gregory, Jörg Strempfer, Daniel Weinstock, Jacob Ruf, Yifei Sun, Hari Nair, Nathaniel J. Schreiber, Darrell G. Schlom, Kyle M. Shen, Andrej Singer

Abstract: In its ground state, RuO2 was long thought to be an ordinary metallic paramagnet. Recent neutron and x-ray diffraction revealed that bulk RuO2 is an antiferromagnet (AFM) with TN above 300 K. Furthermore, epitaxial strain induces novel superconductivity in thin films of RuO2 below 2 K. Here, we present a resonant elastic x-ray scattering (REXS) study at the Ru L2 edge of the strained RuO2 films ex… ▽ More In its ground state, RuO2 was long thought to be an ordinary metallic paramagnet. Recent neutron and x-ray diffraction revealed that bulk RuO2 is an antiferromagnet (AFM) with TN above 300 K. Furthermore, epitaxial strain induces novel superconductivity in thin films of RuO2 below 2 K. Here, we present a resonant elastic x-ray scattering (REXS) study at the Ru L2 edge of the strained RuO2 films exhibiting the strain-induced superconductivity. We observe an azimuthal modulation of the 100 Bragg peak consistent with canted AFM found in bulk. Most notably, in the strained films displaying novel superconductivity, we observe a ~1 eV shift of the Ru eg orbitals to a higher energy. The energy shift is smaller in thicker, relaxed films and films with a different strain direction. Our results provide further evidence of the utility of epitaxial strain as a tuning parameter in complex oxides. △ Less

Submitted 18 January, 2022; originally announced January 2022.

Comments: 20 pages, 3 main figures, 3 supplementary figures

Auction Throttling and Causal Inference of Online Advertising Effects

Authors: George Gui, Harikesh Nair, Fengshi Niu

Abstract: Causally identifying the effect of digital advertising is challenging, because experimentation is expensive, and observational data lacks random variation. This paper identifies a pervasive source of naturally occurring, quasi-experimental variation in user-level ad-exposure in digital advertising campaigns. It shows how this variation can be utilized by ad-publishers to identify the causal effect… ▽ More Causally identifying the effect of digital advertising is challenging, because experimentation is expensive, and observational data lacks random variation. This paper identifies a pervasive source of naturally occurring, quasi-experimental variation in user-level ad-exposure in digital advertising campaigns. It shows how this variation can be utilized by ad-publishers to identify the causal effect of advertising campaigns. The variation pertains to auction throttling, a probabilistic method of budget pacing that is widely used to spread an ad-campaign`s budget over its deployed duration, so that the campaign`s budget is not exceeded or overly concentrated in any one period. The throttling mechanism is implemented by computing a participation probability based on the campaign`s budget spending rate and then including the campaign in a random subset of available ad-auctions each period according to this probability. We show that access to logged-participation probabilities enables identifying the local average treatment effect (LATE) in the ad-campaign. We present a new estimator that leverages this identification strategy and outline a bootstrap procedure for quantifying its variability. We apply our method to real-world ad-campaign data from an e-commerce advertising platform, which uses such throttling for budget pacing. We show our estimate is statistically different from estimates derived using other standard observational methods such as OLS and two-stage least squares estimators. Our estimated conversion lift is 110%, a more plausible number than 600%, the conversion lifts estimated using naive observational methods. △ Less

Submitted 16 February, 2022; v1 submitted 30 December, 2021; originally announced December 2021.

Stochastic MPC with Multi-modal Predictions for Traffic Intersections

Authors: Siddharth H. Nair, Vijay Govindarajan, Theresa Lin, Chris Meissen, H. Eric Tseng, Francesco Borrelli

Abstract: We propose a Stochastic MPC (SMPC) formulation for autonomous driving at traffic intersections which incorporates multi-modal predictions of surrounding vehicles for collision avoidance constraints. The multi-modal predictions are obtained with Gaussian Mixture Models (GMM) and constraints are formulated as chance-constraints. Our main theoretical contribution is a SMPC formulation that optimizes… ▽ More We propose a Stochastic MPC (SMPC) formulation for autonomous driving at traffic intersections which incorporates multi-modal predictions of surrounding vehicles for collision avoidance constraints. The multi-modal predictions are obtained with Gaussian Mixture Models (GMM) and constraints are formulated as chance-constraints. Our main theoretical contribution is a SMPC formulation that optimizes over a novel feedback policy class designed to exploit additional structure in the GMM predictions, and that is amenable to convex programming. The use of feedback policies for prediction is motivated by the need for reduced conservatism in handling multi-modal predictions of the surrounding vehicles, especially prevalent in traffic intersection scenarios. We evaluate our algorithm along axes of mobility, comfort, conservatism and computational efficiency at a simulated intersection in CARLA. Our simulations use a kinematic bicycle model and multimodal predictions trained on a subset of the Lyft Level 5 prediction dataset. To demonstrate the impact of optimizing over feedback policies, we compare our algorithm with two SMPC baselines that handle multi-modal collision avoidance chance constraints by optimizing over open-loop sequences. △ Less

Submitted 25 February, 2022; v1 submitted 20 September, 2021; originally announced September 2021.

Comments: Extended version of ITSC 2022 submission

Tilted spin current generated by the collinear antiferromagnet RuO2

Authors: Arnab Bose, Nathaniel J. Schreiber, Rakshit Jain, Ding-Fu Shao, Hari P. Nair, Jiaxin Sun, Xiyue S. Zhang, David A. Muller, Evgeny Y. Tsymbal, Darrell G. Schlom, Daniel C. Ralph

Abstract: We report measurements demonstrating that when the Neel vector of the collinear antiferromagnet RuO2 is appropriately canted relative to the sample plane, the antiferromagnet generates a substantial out of plane damping-like torque. The measurements are in good accord with predictions that when an electric field, E is applied to the spin split band structure of RuO2 it can cause a strong transvers… ▽ More We report measurements demonstrating that when the Neel vector of the collinear antiferromagnet RuO2 is appropriately canted relative to the sample plane, the antiferromagnet generates a substantial out of plane damping-like torque. The measurements are in good accord with predictions that when an electric field, E is applied to the spin split band structure of RuO2 it can cause a strong transverse spin current even in the absence of spin-orbit coupling. This produces characteristic changes in all three components of the E induced torque vector as a function of the angle of E relative to the crystal axes, corresponding to a spin current with a well defined tilted spin orientation s approximately (but not exactly) parallel to the Neel vector, flowing perpendicular to both E and S. This angular dependence is the signature of an antiferromagnetic spin Hall effect with symmetries that are distinct from other mechanisms of spin-current generation reported in antiferromagnetic or ferromagnetic materials. △ Less

Submitted 20 August, 2021; originally announced August 2021.

Journal ref: Nature Electronics 5, 267 (2022)

Blending Advertising with Organic Content in E-Commerce: A Virtual Bids Optimization Approach

Authors: Carlos Carrion, Zenan Wang, Harikesh Nair, Xianghong Luo, Yulin Lei, Xiliang Lin, Wenlong Chen, Qiyu Hu, Changping Peng, Yongjun Bao, Weipeng Yan

Abstract: In e-commerce platforms, sponsored and non-sponsored content are jointly displayed to users and both may interactively influence their engagement behavior. The former content helps advertisers achieve their marketing goals and provides a stream of ad revenue to the platform. The latter content contributes to users' engagement with the platform, which is key to its long-term health. A burning issue… ▽ More In e-commerce platforms, sponsored and non-sponsored content are jointly displayed to users and both may interactively influence their engagement behavior. The former content helps advertisers achieve their marketing goals and provides a stream of ad revenue to the platform. The latter content contributes to users' engagement with the platform, which is key to its long-term health. A burning issue for e-commerce platform design is how to blend advertising with content in a way that respects these interactions and balances these multiple business objectives. This paper describes a system developed for this purpose in the context of blending personalized sponsored content with non-sponsored content on the product detail pages of JD.COM, an e-commerce company. This system has three key features: (1) Optimization of multiple competing business objectives through a new virtual bids approach and the expressiveness of the latent, implicit valuation of the platform for the multiple objectives via these virtual bids. (2) Modeling of users' click behavior as a function of their characteristics, the individual characteristics of each sponsored content and the influence exerted by other sponsored and non-sponsored content displayed alongside through a deep learning approach; (3) Consideration of externalities in the allocation of ads, thereby making it directly compatible with a Vickrey-Clarke-Groves (VCG) auction scheme for the computation of payments in the presence of these externalities. The system is currently deployed and serving all traffic through JD.COM's mobile application. Experiments demonstrating the performance and advantages of the system are presented. △ Less

Submitted 27 May, 2021; originally announced May 2021.

A Microarchitecture Implementation Framework for Online Learning with Temporal Neural Networks

Authors: Harideep Nair, John Paul Shen, James E. Smith

Abstract: Temporal Neural Networks (TNNs) are spiking neural networks that use time as a resource to represent and process information, similar to the mammalian neocortex. In contrast to compute-intensive deep neural networks that employ separate training and inference phases, TNNs are capable of extremely efficient online incremental/continual learning and are excellent candidates for building edge-native… ▽ More Temporal Neural Networks (TNNs) are spiking neural networks that use time as a resource to represent and process information, similar to the mammalian neocortex. In contrast to compute-intensive deep neural networks that employ separate training and inference phases, TNNs are capable of extremely efficient online incremental/continual learning and are excellent candidates for building edge-native sensory processing units. This work proposes a microarchitecture framework for implementing TNNs using standard CMOS. Gate-level implementations of three key building blocks are presented: 1) multi-synapse neurons, 2) multi-neuron columns, and 3) unsupervised and supervised online learning algorithms based on Spike Timing Dependent Plasticity (STDP). The proposed microarchitecture is embodied in a set of characteristic scaling equations for assessing the gate count, area, delay and power for any TNN design. Post-synthesis results (in 45nm CMOS) for the proposed designs are presented, and their online incremental learning capability is demonstrated. △ Less

Submitted 2 June, 2021; v1 submitted 27 May, 2021; originally announced May 2021.

Comments: To be published in ISVLSI 2021. arXiv admin note: substantial text overlap with arXiv:2009.00457

Journal ref: 2021 IEEE Computer Society Annual Symposium on VLSI (ISVLSI), 2021, pp. 266-271

Low Temperature Domain Wall Freezing and Non-Equilibrium Dynamics in the Transverse-Field Ising Model Material CoNb$_2$O$_6$

Authors: C. L. Sarkis, S. Säubert, V. Williams, E. S. Choi, T. R. Reeder, H. S. Nair, K. A. Ross

Abstract: CoNb$_2$O$_6$ is a rare realization of the transverse field Ising model (TFIM), making it a useful tool for studying both equilibrium and non-equilibrium many-body quantum physics. Despite a large body of work dedicated to characterizing this material, details of the ordered states in the presence of relatively weak transverse fields have not been discussed in detail. Here, we present a detailed s… ▽ More CoNb$_2$O$_6$ is a rare realization of the transverse field Ising model (TFIM), making it a useful tool for studying both equilibrium and non-equilibrium many-body quantum physics. Despite a large body of work dedicated to characterizing this material, details of the ordered states in the presence of relatively weak transverse fields have not been discussed in detail. Here, we present a detailed study of CoNb$_2$O$_6$ via ac susceptibility measurements in order to further characterize its low temperature behavior in the presence of a transverse field. Specifically, we call attention to an unconventional freezing transition in zero-field below T$_F$ = 1.2~K, occurring \emph{within} the well-known commensurate antiferromagnetic (CAFM) state that onsets at $T_{N2}$ = 1.9~K. We performed a series of transverse-field quenches into this frozen state, which resulted in a slowly relaxing susceptibility, $χ^{\prime}(t)$, that followed a logarithmic decay within the time range measured. We discuss the frozen state in the context of the freezing of previously discussed "free" chains arising from domain walls between the four degenerate sublattices of the CAFM state. We also attempted to observe Kibble-Zurek scaling by quenching the transverse field into the frozen state at different rates. This produced a null result; the behavior can be fully explained by coarsening of domains over the timescale of the quenches. The absence of a clear Kibble-Zurek scaling is itself surprising, given the proposed ubiquity of the phenomenon for general second order phase transitions. △ Less

Submitted 6 May, 2021; originally announced May 2021.

Quantum Oscillations and the Quasiparticle Properties of Thin Film Sr$_2$RuO$_4$

Authors: Yawen Fang, Hari P. Nair, Ludi Miao, Berit Goodge, Nathaniel J. Schreiber, Jacob P. Ruf, Lena F. Kourkoutis, Kyle M. Shen, Darrell G. Schlom, B. J. Ramshaw

Abstract: We measure the Shubnikov-de Haas effect in thin-film Sr$_2$RuO$_4$ grown on an (LaAlO$_3$)$_{0.29}$-(SrAl$_{1/2}$Ta$_{1/2}$O$_3$)$_{0.71}$ (LSAT) substrate. We detect all three known Fermi surfaces and extract the Fermi surface volumes, cyclotron effective masses, and quantum lifetimes. We show that the electronic structure is nearly identical to that of single-crystal Sr$_2$RuO$_4$, and that the… ▽ More We measure the Shubnikov-de Haas effect in thin-film Sr$_2$RuO$_4$ grown on an (LaAlO$_3$)$_{0.29}$-(SrAl$_{1/2}$Ta$_{1/2}$O$_3$)$_{0.71}$ (LSAT) substrate. We detect all three known Fermi surfaces and extract the Fermi surface volumes, cyclotron effective masses, and quantum lifetimes. We show that the electronic structure is nearly identical to that of single-crystal Sr$_2$RuO$_4$, and that the quasiparticle lifetime is consistent with the Tc of comparably clean, single-crystal Sr$_2$RuO$_4$. Unlike single-crystal Sr$_2$RuO$_4$, where the quantum and transport lifetimes are roughly equal, we find that the transport lifetime is $1.3\pm0.1$ times longer than the quantum lifetime. This suggests that extended (rather than point) defects may be the dominant source of quasiparticle scattering in these films. To test this idea, we perform cross-sectional STEM and find that out-of-phase boundaries extending the entire thickness of the film occur with a density that is consistent with the quantum mean free path. The long quasiparticle lifetimes make these films ideal for studying the unconventional superconducting state in Sr$_2$RuO$_4$ through the fabrication of devices -- such as planar tunnel junctions and SQUIDs. △ Less

Submitted 2 August, 2021; v1 submitted 28 February, 2021; originally announced March 2021.

Comments: 12 pages, 8 figures

Journal ref: Phys. Rev. B 104, 045152 (2021)

Unsupervised Clustering of Time Series Signals using Neuromorphic Energy-Efficient Temporal Neural Networks

Authors: Shreyas Chaudhari, Harideep Nair, José M. F. Moura, John Paul Shen

Abstract: Unsupervised time series clustering is a challenging problem with diverse industrial applications such as anomaly detection, bio-wearables, etc. These applications typically involve small, low-power devices on the edge that collect and process real-time sensory signals. State-of-the-art time-series clustering methods perform some form of loss minimization that is extremely computationally intensiv… ▽ More Unsupervised time series clustering is a challenging problem with diverse industrial applications such as anomaly detection, bio-wearables, etc. These applications typically involve small, low-power devices on the edge that collect and process real-time sensory signals. State-of-the-art time-series clustering methods perform some form of loss minimization that is extremely computationally intensive from the perspective of edge devices. In this work, we propose a neuromorphic approach to unsupervised time series clustering based on Temporal Neural Networks that is capable of ultra low-power, continuous online learning. We demonstrate its clustering performance on a subset of UCR Time Series Archive datasets. Our results show that the proposed approach either outperforms or performs similarly to most of the existing algorithms while being far more amenable for efficient hardware implementation. Our hardware assessment analysis shows that in 7 nm CMOS the proposed architecture, on average, consumes only about 0.005 mm^2 die area and 22 uW power and can process each signal with about 5 ns latency. △ Less

Submitted 18 February, 2021; originally announced February 2021.

Comments: Accepted for publication at ICASSP 2021

Persephone: A Pluto-System Orbiter and Kuiper Belt Explorer

Authors: Carly Howett, Stuart Robbins, Bryan J. Holler, Amanda Hendrix, Karl Fielhauer, Mark Perry, Fazle Siddique, Clint Apland, James Leary, S. Alan Stern, Heather Elliott, Francis Nimmo, Simon B. Porter, Silvia Protopapa, Kelsi N. Singer, Orenthal J. Tucker, Anne J. Verbiscer, Bruce Andrews, Stewart Bushman, Adam Crifasi, Doug Crowley, Clint Edwards, Carolyn M. Ernst, Blair Fonville, David Frankford , et al. (18 additional authors not shown)

Abstract: Persephone is a NASA concept mission study that addresses key questions raised by New Horizons' encounters with Kuiper Belt objects (KBOs), with arguably the most important being "Does Pluto have a subsurface ocean?". More broadly, Persephone would answer four significant science questions: (1) What are the internal structures of Pluto and Charon? (2) How have the surfaces and atmospheres in the P… ▽ More Persephone is a NASA concept mission study that addresses key questions raised by New Horizons' encounters with Kuiper Belt objects (KBOs), with arguably the most important being "Does Pluto have a subsurface ocean?". More broadly, Persephone would answer four significant science questions: (1) What are the internal structures of Pluto and Charon? (2) How have the surfaces and atmospheres in the Pluto system evolved? (3) How has the KBO population evolved? (4) What are the particles and magnetic field environments of the Kuiper Belt? To answer these questions, Persephone has a comprehensive payload, and would both orbit within the Pluto system and encounter other KBOs. The nominal mission is 30.7 years long, with launch in 2031 on a Space Launch System (SLS) Block 2 rocket with a Centaur kick stage, followed by a 27.6 year cruise powered by existing radioisotope electric propulsion (REP) and a Jupiter gravity assist to reach Pluto in 2058. En route to Pluto, Persephone would have one 50- to 100-km-class KBO encounter before starting a 3.1 Earth-year orbital campaign of the Pluto system. The mission also includes the potential for an 8-year extended mission, which would enable the exploration of another KBO in the 100- to 150-km-size class. The mission payload includes 11 instruments: Panchromatic and Color High-Resolution Imager; Low-Light Camera; Ultra-Violet Spectrometer; Near-Infrared (IR) Spectrometer; Thermal IR Camera; Radio Frequency Spectrometer; Mass Spectrometer; Altimeter; Sounding Radar; Magnetometer; and Plasma Spectrometer. The nominal cost of this mission is $3.0B, making it a large strategic science mission. △ Less

Submitted 16 February, 2021; originally announced February 2021.

Separated transport relaxation scales and interband scattering in SrRuO$_3$, CaRuO$_3$, and Sr$_2$RuO$_4$ thin films

Authors: Youcheng Wang, H. P. Nair, N. J. Schreiber, J. P. Ruf, Bing Cheng, D. G. Schlom, K. M. Shen, N. P. Armitage

Abstract: The anomalous charge transport observed in some strongly correlated metals raises questions as to the universal applicability of Landau Fermi liquid theory. The coherence temperature $T_{FL}$ for normal metals is usually taken to be the temperature below which $T^2$ is observed in the resistivity. Below this temperature, a Fermi liquid with well-defined quasiparticles is expected. However, metalli… ▽ More The anomalous charge transport observed in some strongly correlated metals raises questions as to the universal applicability of Landau Fermi liquid theory. The coherence temperature $T_{FL}$ for normal metals is usually taken to be the temperature below which $T^2$ is observed in the resistivity. Below this temperature, a Fermi liquid with well-defined quasiparticles is expected. However, metallic ruthenates in the Ruddlesden-Popper family, frequently show non-Drude low-energy optical conductivity and unusual $ω/T$ scaling, despite the frequent observation of $T^2$ dc resistivity. Herein we report time-domain THz spectroscopy measurements of several different high-quality metallic ruthenate thin films and show that the optical conductivity can be interpreted in more conventional terms. In all materials, the conductivity has a two-Drude peak lineshape at low temperature and a crossover to a one-Drude peak lineshape at higher temperatures. The two-component low-temperature conductivity is indicative of two well-separated current relaxation rates for different conduction channels. We discuss three particular possibilities for the separation of rates: (a) Strongly energy-dependent inelastic scattering; (b) an almost-conserved pseudomomentum operator that overlaps with the current, giving rise to the narrower Drude peak; (c) the presence of multiple conduction channels that undergoes a crossover to stronger interband scattering at higher temperatures. None of these scenarios require the existence of exotic quasiparticles. The results may give insight into the possible significance of Hund's coupling in determining interband coupling in these materials. Our results also show a route towards understanding the violation of Matthiessen's rule in this class of materials and deviations from the "Gurzhi" scaling relations in Fermi liquids. △ Less

Submitted 23 December, 2020; originally announced December 2020.

Comments: 8 figures. No Supplemental Materials

Journal ref: Phys. Rev. B 103, 205109 (2021)

Mapping the Space of Chemical Reactions Using Attention-Based Neural Networks

Authors: Philippe Schwaller, Daniel Probst, Alain C. Vaucher, Vishnu H. Nair, David Kreutter, Teodoro Laino, Jean-Louis Reymond

Abstract: Organic reactions are usually assigned to classes containing reactions with similar reagents and mechanisms. Reaction classes facilitate the communication of complex concepts and efficient navigation through chemical reaction space. However, the classification process is a tedious task. It requires the identification of the corresponding reaction class template via annotation of the number of mole… ▽ More Organic reactions are usually assigned to classes containing reactions with similar reagents and mechanisms. Reaction classes facilitate the communication of complex concepts and efficient navigation through chemical reaction space. However, the classification process is a tedious task. It requires the identification of the corresponding reaction class template via annotation of the number of molecules in the reactions, the reaction center, and the distinction between reactants and reagents. This work shows that transformer-based models can infer reaction classes from non-annotated, simple text-based representations of chemical reactions. Our best model reaches a classification accuracy of 98.2%. We also show that the learned representations can be used as reaction fingerprints that capture fine-grained differences between reaction classes better than traditional reaction fingerprints. The insights into chemical reaction space enabled by our learned fingerprints are illustrated by an interactive reaction atlas providing visual clustering and similarity searching. △ Less

Submitted 9 December, 2020; originally announced December 2020.

Comments: https://rxn4chemistry.github.io/rxnfp/

A Custom 7nm CMOS Standard Cell Library for Implementing TNN-based Neuromorphic Processors

Authors: Harideep Nair, Prabhu Vellaisamy, Santha Bhasuthkar, John Paul Shen

Abstract: A set of highly-optimized custom macro extensions is developed for a 7nm CMOS cell library for implementing Temporal Neural Networks (TNNs) that can mimic brain-like sensory processing with extreme energy efficiency. A TNN prototype (13,750 neurons and 315,000 synapses) for MNIST requires only 1.56mm2 die area and consumes only 1.69mW. A set of highly-optimized custom macro extensions is developed for a 7nm CMOS cell library for implementing Temporal Neural Networks (TNNs) that can mimic brain-like sensory processing with extreme energy efficiency. A TNN prototype (13,750 neurons and 315,000 synapses) for MNIST requires only 1.56mm2 die area and consumes only 1.69mW. △ Less

Submitted 4 June, 2021; v1 submitted 9 December, 2020; originally announced December 2020.

Comments: This work is dated and will be superseded by a forthcoming work

Inhomogeneous ferromagnetism mimics signatures of the topological Hall effect in SrRuO$_3$ films

Authors: Gideok Kim, K. Son, Y. E. Suyolcu, L. Miao, N. J. Schreiber, H. P. Nair, D. Putzky, M. Minola, G. Christiani, P. A. van Aken, K. M. Shen, D. G. Schlom, G. Logvenov, B. Keimer

Abstract: Topological transport phenomena in magnetic materials are a major topic of current condensed matter research. One of the most widely studied phenomena is the ``topological Hall effect'' (THE), which is generated via spin-orbit interactions between conduction electrons and topological spin textures such as skyrmions. We report a comprehensive set of Hall effect and magnetization measurements on epi… ▽ More Topological transport phenomena in magnetic materials are a major topic of current condensed matter research. One of the most widely studied phenomena is the ``topological Hall effect'' (THE), which is generated via spin-orbit interactions between conduction electrons and topological spin textures such as skyrmions. We report a comprehensive set of Hall effect and magnetization measurements on epitaxial films of the prototypical ferromagnetic metal SrRuO$_3$ the magnetic and transport properties of which were systematically modulated by varying the concentration of Ru vacancies. We observe Hall effect anomalies that closely resemble signatures of the THE, but a quantitative analysis demonstrates that they result from inhomogeneities in the ferromagnetic magnetization caused by a non-random distribution of Ru vacancies. As such inhomogeneities are difficult to avoid and are rarely characterized independently, our results call into question the identification of topological spin textures in numerous prior transport studies of quantum materials, heterostructures, and devices. Firm conclusions regarding the presence of such textures must meet stringent conditions such as probes that couple directly to the non-collinear magnetization on the atomic scale. △ Less

Submitted 19 October, 2020; originally announced October 2020.

Journal ref: Physical Review Materials 4, 104410 (2020)

Comparison Lift: Bandit-based Experimentation System for Online Advertising

Authors: Tong Geng, Xiliang Lin, Harikesh S. Nair, Jun Hao, Bin Xiang, Shurui Fan

Abstract: Comparison Lift is an experimentation-as-a-service (EaaS) application for testing online advertising audiences and creatives at JD.com. Unlike many other EaaS tools that focus primarily on fixed sample A/B testing, Comparison Lift deploys a custom bandit-based experimentation algorithm. The advantages of the bandit-based approach are two-fold. First, it aligns the randomization induced in the test… ▽ More Comparison Lift is an experimentation-as-a-service (EaaS) application for testing online advertising audiences and creatives at JD.com. Unlike many other EaaS tools that focus primarily on fixed sample A/B testing, Comparison Lift deploys a custom bandit-based experimentation algorithm. The advantages of the bandit-based approach are two-fold. First, it aligns the randomization induced in the test with the advertiser's goals from testing. Second, by adapting experimental design to information acquired during the test, it reduces substantially the cost of experimentation to the advertiser. Since launch in May 2019, Comparison Lift has been utilized in over 1,500 experiments. We estimate that utilization of the product has helped increase click-through rates of participating advertising campaigns by 46% on average. We estimate that the adaptive design in the product has generated 27% more clicks on average during testing compared to a fixed sample A/B design. Both suggest significant value generation and cost savings to advertisers from the product. △ Less

Submitted 16 September, 2020; originally announced September 2020.

Direct CMOS Implementation of Neuromorphic Temporal Neural Networks for Sensory Processing

Authors: Harideep Nair, John Paul Shen, James E. Smith

Abstract: Temporal Neural Networks (TNNs) use time as a resource to represent and process information, mimicking the behavior of the mammalian neocortex. This work focuses on implementing TNNs using off-the-shelf digital CMOS technology. A microarchitecture framework is introduced with a hierarchy of building blocks including: multi-neuron columns, multi-column layers, and multi-layer TNNs. We present the d… ▽ More Temporal Neural Networks (TNNs) use time as a resource to represent and process information, mimicking the behavior of the mammalian neocortex. This work focuses on implementing TNNs using off-the-shelf digital CMOS technology. A microarchitecture framework is introduced with a hierarchy of building blocks including: multi-neuron columns, multi-column layers, and multi-layer TNNs. We present the direct CMOS gate-level implementation of the multi-neuron column model as the key building block for TNNs. Post-synthesis results are obtained using Synopsys tools and the 45 nm CMOS standard cell library. The TNN microarchitecture framework is embodied in a set of characteristic equations for assessing the total gate count, die area, compute time, and power consumption for any TNN design. We develop a multi-layer TNN prototype of 32M gates. In 7 nm CMOS process, it consumes only 1.54 mm^2 die area and 7.26 mW power and can process 28x28 images at 107M FPS (9.34 ns per image). We evaluate the prototype's performance and complexity relative to a recent state-of-the-art TNN model. △ Less

Submitted 27 August, 2020; originally announced September 2020.

Comments: Submission Under Review for an IEEE Conference