×

Arrangement of nearby minima and saddles in the mixed spherical energy landscapes. (English) Zbl 07908479

Summary: The mixed spherical models were recently found to violate long-held assumptions about mean-field glassy dynamics. In particular, the threshold energy, where most stationary points are marginal and that in the simpler pure models attracts long-time dynamics, seems to lose significance. Here, we compute the typical distribution of stationary points relative to each other in mixed models with a replica symmetric complexity. We examine the stability of nearby points, accounting for the presence of an isolated eigenvalue in their spectrum due to their proximity. Despite finding rich structure not present in the pure models, we find nothing that distinguishes the points that do attract the dynamics. Instead, we find new geometric significance of the old threshold energy, and invalidate pictures of the arrangement of most marginal inherent states into a continuous manifold.

MSC:

82Dxx Applications of statistical mechanics to specific types of physical systems
82Bxx Equilibrium statistical mechanics
57-XX Manifolds and cell complexes

References:

[1] D. L. Stein and C. M. Newman, Broken ergodicity and the geometry of rugged landscapes, Phys. Rev. E 51, 5228 (1995), doi:10.1103/physreve.51.5228. · doi:10.1103/physreve.51.5228
[2] F. Krzakala and J. Kurchan, Landscape analysis of constraint satisfaction problems, Phys. Rev. E 76, 021122 (2007), doi:10.1103/physreve.76.021122. · doi:10.1103/physreve.76.021122
[3] A. Altieri, F. Roy, C. Cammarota and G. Biroli, Properties of equilibria and glassy phases of the random Lotka-Volterra model with demographic noise, Phys. Rev. Lett. 126, 258301 (2021), doi:10.1103/physrevlett.126.258301. · doi:10.1103/physrevlett.126.258301
[4] N. Yang, C. Tang and Y. Tu, Stochastic gradient descent introduces an effective landscape-dependent regularization favoring flat solutions, Phys. Rev. Lett. 130, 237101 (2023), doi:10.1103/physrevlett.130.237101. · doi:10.1103/physrevlett.130.237101
[5] A. Cavagna, Fragile vs. strong liquids: A saddles-ruled scenario, Europhys. Lett. 53, 490 (2001), doi:10.1209/epl/i2001-00179-4. · doi:10.1209/epl/i2001-00179-4
[6] F. H. Stillinger and P. G. Debenedetti, Glass transition thermodynamics and kinetics, Annu. Rev. Condens. Matter Phys. 4, 263 (2013), doi:10.1146/annurev-conmatphys-030212-184329. · doi:10.1146/annurev-conmatphys-030212-184329
[7] T. R. Kirkpatrick and D. Thirumalai, Colloquium: Random first order transi-tion theory concepts in biology and physics, Rev. Mod. Phys. 87, 183 (2015), doi:10.1103/revmodphys.87.183. · doi:10.1103/revmodphys.87.183
[8] T. Castellani and A. Cavagna, Spin-glass theory for pedestrians, J. Stat. Mech.: Theory Exp. P05012 (2005), doi:10.1088/1742-5468/2005/05/p05012. · Zbl 1456.82490 · doi:10.1088/1742-5468/2005/05/p05012
[9] G. Biroli, Dynamical TAP approach to mean field glassy systems, J. Phys. A: Math. Gen. 32, 8365 (1999), doi:10.1088/0305-4470/32/48/301. · Zbl 0955.82031 · doi:10.1088/0305-4470/32/48/301
[10] M. Sellke, The threshold energy of low temperature Langevin dynamics for pure spherical spin glasses, (arXiv preprint) doi:10.48550/arXiv.2305.07956. · doi:10.48550/arXiv.2305.07956
[11] G. Folena, S. Franz and F. Ricci-Tersenghi, Rethinking mean-field glassy dynamics and its relation with the energy landscape: The surprising case of the spherical mixed p-spin model, Phys. Rev. X 10, 031045 (2020), doi:10.1103/PhysRevX.10.031045. · doi:10.1103/PhysRevX.10.031045
[12] G. Folena, S. Franz and F. Ricci-Tersenghi, Gradient descent dynamics in the mixed p-spin spherical model: Finite-size simulations and comparison with mean-field integration, J. Stat. Mech.: Theory Exp. 033302 (2021), doi:10.1088/1742-5468/abe29f. · Zbl 1504.82051 · doi:10.1088/1742-5468/abe29f
[13] V. Ros, G. Biroli and C. Cammarota, Complexity of energy barriers in mean-field glassy systems, Europhys. Lett. 126, 20003 (2019), doi:10.1209/0295-5075/126/20003. · doi:10.1209/0295-5075/126/20003
[14] J. Kurchan and L. Laloux, Phase space geometry and slow dynamics, J. Phys. A: Math. Gen. 29, 1929 (1996), doi:10.1088/0305-4470/29/9/009. · Zbl 0900.70205 · doi:10.1088/0305-4470/29/9/009
[15] T. R. Kirkpatrick and D. Thirumalai, p-spin-interaction spin-glass models: Con-nections with the structural glass problem, Phys. Rev. B 36, 5388 (1987), doi:10.1103/physrevb.36.5388. · doi:10.1103/physrevb.36.5388
[16] A. Crisanti and H.-J. Sommers, The spherical p-spin interaction spin glass model: The statics, Z. Phys. B -Condens. Matter 87, 341 (1992), doi:10.1007/bf01309287. · doi:10.1007/bf01309287
[17] A. Crisanti and L. Leuzzi, Spherical 2 + p spin-glass model: An exactly solv-able model for glass to spin-glass transition, Phys. Rev. Lett. 93, 217203 (2004), doi:10.1103/physrevlett.93.217203. · doi:10.1103/physrevlett.93.217203
[18] A. Crisanti and L. Leuzzi, Spherical 2 + p spin-glass model: An analytically solv-able model with a glass-to-glass transition, Phys. Rev. B 73, 014412 (2006), doi:10.1103/physrevb.73.014412. · doi:10.1103/physrevb.73.014412
[19] J. Kent-Dobias, When is the average number of saddle points typical?, Europhys. Lett. 143, 61003 (2023), doi:10.1209/0295-5075/acf521. · doi:10.1209/0295-5075/acf521
[20] G. B. Arous, E. Subag and O. Zeitouni, Geometry and temperature chaos in mixed spherical spin glasses at low temperature: The perturbative regime, Comm. Pure Appl. Math. 73, 1732 (2019), doi:10.1002/cpa.21875. · Zbl 1453.82089 · doi:10.1002/cpa.21875
[21] M. Audin and M. Damian, Morse theory and Floer homology, Springer, London, UK, ISBN 9781447154952 (2014), doi:10.1007/978-1-4471-5496-9. · Zbl 1281.57001 · doi:10.1007/978-1-4471-5496-9
[22] G. Folena and F. Zamponi, On weak ergodicity breaking in mean-field spin glasses, SciPost Phys. 15, 109 (2023), doi:10.21468/scipostphys.15.3.109. · Zbl 07905044 · doi:10.21468/scipostphys.15.3.109
[23] M. Kac, On the average number of real roots of a random algebraic equation, Bull. Am. Math. Soc. 49, 314 (1943). · Zbl 0060.28602
[24] S. O. Rice, Mathematical analysis of random noise, Bell Syst. Tech. J. 23, 282 (1944), doi:10.1002/j.1538-7305.1944.tb00874.x. · Zbl 0063.06485 · doi:10.1002/j.1538-7305.1944.tb00874.x
[25] A. Cavagna, I. Giardina and G. Parisi, Stationary points of the Thouless-Anderson-Palmer free energy, Phys. Rev. B 57, 11251 (1998), doi:10.1103/physrevb.57.11251. · doi:10.1103/physrevb.57.11251
[26] Y. V. Fyodorov, H.-J. Sommers and I. Williams, Density of stationary points in a high di-mensional random energy landscape and the onset of glassy behavior, JETP Lett. 85, 261 (2007), doi:10.1134/s0021364007050098. · doi:10.1134/s0021364007050098
[27] A. J. Bray and D. S. Dean, Statistics of critical points of Gaussian fields on large-dimensional spaces, Phys. Rev. Lett. 98, 150201 (2007), doi:10.1103/physrevlett.98.150201. · doi:10.1103/physrevlett.98.150201
[28] J. Kent-Dobias and J. Kurchan, How to count in hierarchical landscapes: A full solution to mean-field complexity, Phys. Rev. E 107, 064111 (2023), doi:10.1103/PhysRevE.107.064111. · doi:10.1103/PhysRevE.107.064111
[29] H. Ikeda, Bose-Einstein-like condensation of deformed random matrix: A replica approach, J. Stat. Mech.: Theory Exp. 023302 (2023), doi:10.1088/1742-5468/acb7d6. · Zbl 1539.82113 · doi:10.1088/1742-5468/acb7d6
[30] C. Baldassi, C. Borgs, J. T. Chayes, A. Ingrosso, C. Lucibello, L. Saglietti and R. Zecchina, Unreasonable effectiveness of learning neural networks: From accessible states and ro-bust ensembles to basic algorithmic schemes, Proc. Natl. Acad. Sci. 113, E7655 (2016), doi:10.1073/pnas.1608103113. · doi:10.1073/pnas.1608103113
[31] C. Baldassi, C. Lauditi, E. M. Malatesta, G. Perugini and R. Zecchina, Unveiling the structure of wide flat minima in neural networks, Phys. Rev. Lett. 127, 278301 (2021), doi:10.1103/physrevlett.127.278301. · doi:10.1103/physrevlett.127.278301
[32] S. Franz and G. Parisi, Recipes for metastable states in spin glasses, J. Phys. I France 5, 1401 (1995), doi:10.1051/jp1:1995201. · doi:10.1051/jp1:1995201
[33] S. Franz and G. Parisi, Effective potential in glassy systems: Theory and simulations, Phys. A: Stat. Mech. Appl. 261, 317 (1998), doi:10.1016/s0378-4371(98)00315-x. · doi:10.1016/s0378-4371(98)00315-x
This reference list is based on information provided by the publisher or from digital mathematics libraries. Its items are heuristically matched to zbMATH identifiers and may contain data conversion errors. In some cases that data have been complemented/enhanced by data from zbMATH Open. This attempts to reflect the references listed in the original paper as accurately as possible without claiming completeness or a perfect matching.