-
Tilt and Average : Geometric Adjustment of the Last Layer for Recalibration
Authors:
Gyusang Cho,
Chan-Hyun Youn
Abstract:
After the revelation that neural networks tend to produce overconfident predictions, the problem of calibration, which aims to align confidence with accuracy to enhance the reliability of predictions, has gained significant importance. Several solutions based on calibration maps have been proposed to address the problem of recalibrating a trained classifier using additional datasets. In this paper…
▽ More
After the revelation that neural networks tend to produce overconfident predictions, the problem of calibration, which aims to align confidence with accuracy to enhance the reliability of predictions, has gained significant importance. Several solutions based on calibration maps have been proposed to address the problem of recalibrating a trained classifier using additional datasets. In this paper, we offer an algorithm that transforms the weights of the last layer of the classifier, distinct from the calibration-map-based approach. We concentrate on the geometry of the final linear layer, specifically its angular aspect, and adjust the weights of the corresponding layer. We name the method Tilt and Average(\textsc{Tna}), and validate the calibration effect empirically and theoretically. Through this, we demonstrate that our approach, in addition to the existing calibration-map-based techniques, can yield improved calibration performance. Code available : https://github.com/GYYYYYUUUUU/TNA_Angular_Scaling.
△ Less
Submitted 14 June, 2024;
originally announced June 2024.
-
Influence Pathway Discovery on Social Media
Authors:
Xinyi Liu,
Ruijie Wang,
Dachun Sun,
Jinning Li,
Christina Youn,
You Lyu,
Jianyuan Zhan,
Dayou Wu,
Xinhe Xu,
Mingjun Liu,
Xinshuo Lei,
Zhihao Xu,
Yutong Zhang,
Zehao Li,
Qikai Yang,
Tarek Abdelzaher
Abstract:
This paper addresses influence pathway discovery, a key emerging problem in today's online media. We propose a discovery algorithm that leverages recently published work on unsupervised interpretable ideological embedding, a mapping of ideological beliefs (done in a self-supervised fashion) into interpretable low-dimensional spaces. Computing the ideological embedding at scale allows one to analyz…
▽ More
This paper addresses influence pathway discovery, a key emerging problem in today's online media. We propose a discovery algorithm that leverages recently published work on unsupervised interpretable ideological embedding, a mapping of ideological beliefs (done in a self-supervised fashion) into interpretable low-dimensional spaces. Computing the ideological embedding at scale allows one to analyze correlations between the ideological positions of leaders, influencers, news portals, or population segments, deriving potential influence pathways. The work is motivated by the importance of social media as the preeminent means for global interactions and collaborations on today's Internet, as well as their frequent (mis-)use to wield influence that targets social beliefs and attitudes of selected populations. Tools that enable the understanding and mapping of influence propagation through population segments on social media are therefore increasingly important. In this paper, influence is measured by the perceived ideological shift over time that is correlated with influencers' activity. Correlated shifts in ideological embeddings indicate changes, such as swings/switching (among competing ideologies), polarization (depletion of neutral ideological positions), escalation/radicalization (shifts to more extreme versions of the ideology), or unification/cooldown (shifts towards more neutral stances). Case-studies are presented to explore selected influence pathways (i) in a recent French election, (ii) during political discussions in the Philippines, and (iii) for some Russian messaging during the Russia/Ukraine conflict.
△ Less
Submitted 27 September, 2023;
originally announced September 2023.
-
FREEDOM: Target Label & Source Data & Domain Information-Free Multi-Source Domain Adaptation for Unsupervised Personalization
Authors:
Eunju Yang,
Gyusang Cho,
Chan-Hyun Youn
Abstract:
From a service perspective, Multi-Source Domain Adaptation (MSDA) is a promising scenario to adapt a deployed model to a client's dataset. It can provide adaptation without a target label and support the case where a source dataset is constructed from multiple domains. However, it is impractical, wherein its training heavily relies on prior domain information of the multi-source dataset -- how man…
▽ More
From a service perspective, Multi-Source Domain Adaptation (MSDA) is a promising scenario to adapt a deployed model to a client's dataset. It can provide adaptation without a target label and support the case where a source dataset is constructed from multiple domains. However, it is impractical, wherein its training heavily relies on prior domain information of the multi-source dataset -- how many domains exist and the domain label of each data sample. Moreover, MSDA requires both source and target datasets simultaneously (physically), causing storage limitations on the client device or data privacy issues by transferring client data to a server. For a more practical scenario of model adaptation from a service provider's point of view, we relax these constraints and present a novel problem scenario of Three-Free Domain Adaptation, namely TFDA, where 1) target labels, 2) source dataset, and mostly 3) source domain information (domain labels + the number of domains) are unavailable. Under the problem scenario, we propose a practical adaptation framework called FREEDOM. It leverages the power of the generative model, disentangling data into class and style aspects, where the style is defined as the class-independent information from the source data and designed with a nonparametric Bayesian approach. In the adaptation stage, FREEDOM aims to match the source class distribution with the target's under the philosophy that class distribution is consistent even if the style is different; after then, only part of the classification model is deployed as a personalized network. As a result, FREEDOM achieves state-of-the-art or comparable performance even without domain information, with reduced final model size on the target side, independent of the number of source domains.
△ Less
Submitted 4 July, 2023;
originally announced July 2023.
-
Building A Trusted Execution Environment for In-Storage Computing
Authors:
Yuqi Xue,
Luyi Kang,
Weiwei Jia,
Xiaohao Wang,
Jongryool Kim,
Changhwan Youn,
Myeong Joon Kang,
Hyung Jin Lim,
Bruce Jacob,
Jian Huang
Abstract:
In-storage computing with modern solid-state drives (SSDs) enables developers to offload programs from the host to the SSD. It has been proven to be an effective approach to alleviating the I/O bottleneck. To facilitate in-storage computing, many frameworks have been proposed. However, few of them consider security as the priority for in-storage computing. Specifically, since modern SSD controller…
▽ More
In-storage computing with modern solid-state drives (SSDs) enables developers to offload programs from the host to the SSD. It has been proven to be an effective approach to alleviating the I/O bottleneck. To facilitate in-storage computing, many frameworks have been proposed. However, few of them consider security as the priority for in-storage computing. Specifically, since modern SSD controllers do not have a trusted execution environment, an offloaded (malicious) program could steal, modify, and even destroy the data stored in the SSD. In this paper, we first investigate the attacks that could be conducted by offloaded in-storage programs. To defend against these attacks, we build IceClave, a lightweight trusted execution environment for in-storage computing. IceClave enables security isolation between in-storage programs and flash management functions. IceClave also achieves security isolation between in-storage programs and enforces memory encryption and integrity verification of in-storage DRAM with low overhead. To protect data loaded from flash chips, IceClave develops a lightweight data encryption/decryption mechanism in flash controllers. We develop IceClave with a full system simulator and evaluate IceClave with a variety of data-intensive applications. Compared to state-of-the-art in-storage computing approaches, IceClave introduces only 7.6% performance overhead, while enforcing security isolation in the SSD controller with minimal hardware cost. IceClave still keeps the performance benefit of in-storage computing by delivering up to 2.31$\times$ better performance than the conventional host-based trusted computing approach.
△ Less
Submitted 12 May, 2022;
originally announced May 2022.
-
NEMAR: An open access data, tools, and compute resource operating on NeuroElectroMagnetic data
Authors:
Arnaud Delorme,
Dung Truong,
Choonhan Youn,
Subha Sivagnanam,
Kenneth Yoshimoto,
Russell A. Poldrack,
Amit Majumdar,
Scott Makeig
Abstract:
To take advantage of recent and ongoing advances in large-scale computational methods, and to preserve the scientific data created by publicly funded research projects, data archives must be created as well as standards for specifying, identifying, and annotating deposited data. The OpenNeuro.org archive, begun as a repository for magnetic resonance imaging (MRI) data, is such an archive. We prese…
▽ More
To take advantage of recent and ongoing advances in large-scale computational methods, and to preserve the scientific data created by publicly funded research projects, data archives must be created as well as standards for specifying, identifying, and annotating deposited data. The OpenNeuro.org archive, begun as a repository for magnetic resonance imaging (MRI) data, is such an archive. We present a gateway to OpenNeuro for human electrophysiology data (BIDS-formatted EEG and MEG, as well as intracranial data). The NEMAR gateway allows users to visualize electrophysiological data, including time-domain and frequency-domain dynamics time locked to sets of experimental events recorded using BIDS- and HED-formatted data annotation. In addition, NEMAR allows users to process archived EEG data on the XSEDE high-performance resources at SDSC in conjunction with the Neuroscience Gateway (nsgportal.org), a freely available and easy to use portal to leverage high-performance computing resources for neuroscience research.
△ Less
Submitted 4 March, 2022;
originally announced March 2022.
-
IceClave: A Trusted Execution Environment for In-Storage Computing
Authors:
Luyi Kang,
Yuqi Xue,
Weiwei Jia,
Xiaohao Wang,
Jongryool Kim,
Changhwan Youn,
Myeong Joon Kang,
Hyung Jin Lim,
Bruce Jacob,
Jian Huang
Abstract:
In-storage computing with modern solid-state drives (SSDs) enables developers to offload programs from the host to the SSD. It has been proven to be an effective approach to alleviate the I/O bottleneck. To facilitate in-storage computing, many frameworks have been proposed. However, few of them treat the in-storage security as the first citizen. Specifically, since modern SSD controllers do not h…
▽ More
In-storage computing with modern solid-state drives (SSDs) enables developers to offload programs from the host to the SSD. It has been proven to be an effective approach to alleviate the I/O bottleneck. To facilitate in-storage computing, many frameworks have been proposed. However, few of them treat the in-storage security as the first citizen. Specifically, since modern SSD controllers do not have a trusted execution environment, an offloaded (malicious) program could steal, modify, and even destroy the data stored in the SSD. In this paper, we first investigate the attacks that could be conducted by offloaded in-storage programs. To defend against these attacks, we build a lightweight trusted execution environment, named IceClave for in-storage computing. IceClave enables security isolation between in-storage programs and flash management functions that include flash address translation, data access control, and garbage collection, with TrustZone extensions. IceClave also achieves security isolation between in-storage programs by enforcing memory integrity verification of in-storage DRAM with low overhead. To protect data loaded from flash chips, IceClave develops a lightweight data encryption/decryption mechanism in flash controllers. We develop IceClave with a full system simulator. We evaluate IceClave with a variety of data-intensive applications such as databases. Compared to state-of-the-art in-storage computing approaches, IceClave introduces only 7.6% performance overhead, while enforcing security isolation in the SSD controller with minimal hardware cost. IceClave still keeps the performance benefit of in-storage computing by delivering up to 2.31$\times$ better performance than the conventional host-based trusted computing approach.
△ Less
Submitted 7 September, 2021;
originally announced September 2021.
-
AOMD: An Analogy-aware Approach to Offensive Meme Detection on Social Media
Authors:
Lanyu Shang,
Yang Zhang,
Yuheng Zha,
Yingxi Chen,
Christina Youn,
Dong Wang
Abstract:
This paper focuses on an important problem of detecting offensive analogy meme on online social media where the visual content and the texts/captions of the meme together make an analogy to convey the offensive information. Existing offensive meme detection solutions often ignore the implicit relation between the visual and textual contents of the meme and are insufficient to identify the offensiv…
▽ More
This paper focuses on an important problem of detecting offensive analogy meme on online social media where the visual content and the texts/captions of the meme together make an analogy to convey the offensive information. Existing offensive meme detection solutions often ignore the implicit relation between the visual and textual contents of the meme and are insufficient to identify the offensive analogy memes. Two important challenges exist in accurately detecting the offensive analogy memes: i) it is not trivial to capture the analogy that is often implicitly conveyed by a meme; ii) it is also challenging to effectively align the complex analogy across different data modalities in a meme. To address the above challenges, we develop a deep learning based Analogy-aware Offensive Meme Detection (AOMD) framework to learn the implicit analogy from the multi-modal contents of the meme and effectively detect offensive analogy memes. We evaluate AOMD on two real-world datasets from online social media. Evaluation results show that AOMD achieves significant performance gains compared to state-of-the-art baselines by detecting offensive analogy memes more accurately.
△ Less
Submitted 21 June, 2021;
originally announced June 2021.
-
City-Scale Visual Place Recognition with Deep Local Features Based on Multi-Scale Ordered VLAD Pooling
Authors:
Duc Canh Le,
Chan Hyun Youn
Abstract:
Visual place recognition is the task of recognizing a place depicted in an image based on its pure visual appearance without metadata. In visual place recognition, the challenges lie upon not only the changes in lighting conditions, camera viewpoint, and scale but also the characteristic of scene-level images and the distinct features of the area. To resolve these challenges, one must consider bot…
▽ More
Visual place recognition is the task of recognizing a place depicted in an image based on its pure visual appearance without metadata. In visual place recognition, the challenges lie upon not only the changes in lighting conditions, camera viewpoint, and scale but also the characteristic of scene-level images and the distinct features of the area. To resolve these challenges, one must consider both the local discriminativeness and the global semantic context of images. On the other hand, the diversity of the datasets is also particularly important to develop more general models and advance the progress of the field. In this paper, we present a fully-automated system for place recognition at a city-scale based on content-based image retrieval. Our main contributions to the community lie in three aspects. Firstly, we take a comprehensive analysis of visual place recognition and sketch out the unique challenges of the task compared to general image retrieval tasks. Next, we propose yet a simple pooling approach on top of convolutional neural network activations to embed the spatial information into the image representation vector. Finally, we introduce new datasets for place recognition, which are particularly essential for application-based research. Furthermore, throughout extensive experiments, various issues in both image retrieval and place recognition are analyzed and discussed to give some insights into improving the performance of retrieval models in reality.
The dataset used in this paper can be found at https://github.com/canhld94/Daejeon520
△ Less
Submitted 1 May, 2023; v1 submitted 19 September, 2020;
originally announced September 2020.
-
Measurement-Protected Quantum Key Distribution
Authors:
Spiros Kechrimparis,
Heasin Ko,
Young-Ho Ko,
Kap-Joong Kim,
Byung-Seok Choi,
Chahan M. Kropf,
Chun Ju Youn,
Joonwoo Bae
Abstract:
In the distribution of quantum states over a long distance, not only are quantum states corrupted by interactions with an environment but also a measurement setting should be re-aligned such that detection events can be ensured for the resulting states. In this work, we present measurement-protected quantum key distribution where a measurement is protected against the interactions quantum states e…
▽ More
In the distribution of quantum states over a long distance, not only are quantum states corrupted by interactions with an environment but also a measurement setting should be re-aligned such that detection events can be ensured for the resulting states. In this work, we present measurement-protected quantum key distribution where a measurement is protected against the interactions quantum states experience during the transmission, without the verification of a channel. As a result, a receiver does not have to revise the measurement that has been prepared in a noiseless scenario since it would remain ever optimal. The measurement protection is achieved by applications of local unitary transformations before and after the transmission, that leads to a supermap transforming an arbitrary channel to a depolarization one. An experimental demonstration is presented with the polarization encoding on photonic qubits. It is shown that the security bounds for prepare-and-measure protocols can be improved, for instance, errors up to 20.7% can be tolerated in the Bennett-Brassard 1984 protocol.
△ Less
Submitted 2 December, 2019;
originally announced December 2019.
-
Criteria and Approaches for Virtualization on Modern FPGAs
Authors:
Duc-Canh Le,
Chan-Hyun Youn
Abstract:
Modern field programmable gate arrays (FPGAs) can produce high performance in a wide range of applications, and their computational capacity is becoming abundant in personal computers. Regardless of this fact, FPGA virtualization is an emerging research field. Nowadays, challenges of the research area come from not only technical difficulties but also from the ambiguous standards of virtualization…
▽ More
Modern field programmable gate arrays (FPGAs) can produce high performance in a wide range of applications, and their computational capacity is becoming abundant in personal computers. Regardless of this fact, FPGA virtualization is an emerging research field. Nowadays, challenges of the research area come from not only technical difficulties but also from the ambiguous standards of virtualization. In this paper, we introduce novel criteria of FPGA virtualization and discuss several approaches to accomplish those criteria. In addition, we present and describe in detail the specific FPGA virtualization architecture that we developed on Intel Arria 10 FPGA. We evaluate our solution with a combination of applications and microbenchmarks. The result shows that our virtualization solution can provide a full abstraction of FPGA device in both user and developer perspective while maintaining a reasonable performance compared to native FPGA.
△ Less
Submitted 8 April, 2019;
originally announced April 2019.
-
High-speed and high-performance polarization-based quantum key distribution system without side channel effects caused by multiple lasers
Authors:
Heasin Ko,
Byung-Seok Choi,
Joong-Seon Choe,
Kap-Joong Kim,
Jong-Hoi Kim,
Chun Ju Youn
Abstract:
Side channel effects such as temporal disparity and intensity fluctuation of photon pulses caused by random bit generation with multiple laser diodes in high-speed polarization-based BB84 quantum key distribution (QKD) systems can be eliminated by increasing DC bias current condition. However, background photons caused by the spontaneous emission process under high DC bias current degrade the perf…
▽ More
Side channel effects such as temporal disparity and intensity fluctuation of photon pulses caused by random bit generation with multiple laser diodes in high-speed polarization-based BB84 quantum key distribution (QKD) systems can be eliminated by increasing DC bias current condition. However, background photons caused by the spontaneous emission process under high DC bias current degrade the performance of the QKD systems. In this study, we investigated, for the first time, the effects of spontaneously emitted photons on the system performance in a high-speed QKD system at a clock rate of 400 MHz. Also, we further show improvements of system performance without side channel effects by utilizing temporal filtering technique with real-time FPGA signal processing.
△ Less
Submitted 22 November, 2017;
originally announced November 2017.
-
Critical side channel effects in random bit generation with multiple semiconductor lasers in a polarization-based quantum key distribution system
Authors:
Heasin Ko,
Byung-Seok Choi,
Joong-Seon Choe,
Kap-Joong Kim,
Jong-Hoi Kim,
Chun Ju Youn
Abstract:
Most polarization-based BB84 quantum key distribution (QKD) systems utilize multiple lasers to generate one of four polarization quantum states randomly. However, random bit generation with multiple lasers can potentially open critical side channels, which significantly endangers the security of QKD systems. In this paper, we show unnoticed side channels of temporal disparity and intensity fluctua…
▽ More
Most polarization-based BB84 quantum key distribution (QKD) systems utilize multiple lasers to generate one of four polarization quantum states randomly. However, random bit generation with multiple lasers can potentially open critical side channels, which significantly endangers the security of QKD systems. In this paper, we show unnoticed side channels of temporal disparity and intensity fluctuation, which possibly exist in the operation of multiple semiconductor laser diodes. Experimental results show that the side channels can enormously degrade security performance of QKD systems. An important system issue for the improvement of quantum bit error rate (QBER) related with laser driving condition is furtherly addressed with experimental results.
△ Less
Submitted 27 June, 2017;
originally announced June 2017.
-
Enhanced electrocatalytic activity via phase transitions in strongly correlated SrRuO3 thin films
Authors:
Sang A Lee,
Seokjae Oh,
Jae-Yeol Hwang,
Minseok Choi,
Chulmin Youn,
Ji Woong Kim,
Seo Hyoung Chang,
Sungmin Woo,
Jong-Seong Bae,
Sungkyun Park,
Young-Min Kim,
Suyoun Lee,
Taekjib Choi,
Sung Wng Kim,
Woo Seok Choi
Abstract:
Transition metal oxides have been extensively studied and utilized as efficient catalysts. However, the strongly correlated behavior which often results in intriguing emergent phenomena in these materials has been mostly overlooked in understanding the electrochemical activities. Here, we demonstrate a close correlation between the phase transitions and oxygen evolution reaction (OER) in a strongl…
▽ More
Transition metal oxides have been extensively studied and utilized as efficient catalysts. However, the strongly correlated behavior which often results in intriguing emergent phenomena in these materials has been mostly overlooked in understanding the electrochemical activities. Here, we demonstrate a close correlation between the phase transitions and oxygen evolution reaction (OER) in a strongly correlated SrRuO3. By systematically introducing Ru-O vacancies into the single-crystalline SrRuO3 epitaxial thin films, we induced phase transition in crystalline symmetry which resulted in corresponding modification in the electronic structure. The modified electronic structure significantly affect the electrochemical activities, so a 30% decrease in the overpotential for the OER activity was achieved. Our study suggests that a substantial enhancement in the OER activity can be realized even within single material systems, by rational design and engineering of their crystal and electronic structures.
△ Less
Submitted 11 May, 2017;
originally announced May 2017.
-
Hybrid Workflow Policy Management for Heart Disease Identification
Authors:
Dong-Hyun Kim,
Woo-Ram Jung,
Chan-Hyun Youn
Abstract:
As science technology grows, medical application is becoming more complex to solve the physiological problems within expected time. Workflow management systems (WMS) in Grid computing are promising solution to solve the sophisticated problem such as genomic analysis, drug discovery, disease identification, etc. Although existing WMS can provide basic management functionality in Grid environment,…
▽ More
As science technology grows, medical application is becoming more complex to solve the physiological problems within expected time. Workflow management systems (WMS) in Grid computing are promising solution to solve the sophisticated problem such as genomic analysis, drug discovery, disease identification, etc. Although existing WMS can provide basic management functionality in Grid environment, consideration of user requirements such as performance, reliability and interaction with user is missing. In this paper, we propose hybrid workflow management system for heart disease identification and discuss how to guarantee different user requirements according to user SLA. The proposed system is applied to Physio-Grid e-health platform to identify human heart disease with ECG analysis and Virtual Heart Simulation (VHS) workflow applications.
△ Less
Submitted 23 January, 2010;
originally announced January 2010.
