Statistical physics of inference: Thresholds and algorithms
L Zdeborová, F Krzakala - Advances in Physics, 2016 - Taylor & Francis
Many questions of fundamental interest in today's science can be formulated as inference
problems: some partial, or noisy, observations are performed over a set of variables and the …
problems: some partial, or noisy, observations are performed over a set of variables and the …
Mode-coupling theory and the glass transition in supercooled liquids
SP Das - Reviews of modern physics, 2004 - APS
Mode-coupling theory is an approach to the study of complex behavior in the supercooled
liquids which developed from the idea of a nonlinear feedback mechanism. From the …
liquids which developed from the idea of a nonlinear feedback mechanism. From the …
High-dimensional limit theorems for sgd: Effective dynamics and critical scaling
G Ben Arous, R Gheissari… - Advances in Neural …, 2022 - proceedings.neurips.cc
We study the scaling limits of stochastic gradient descent (SGD) with constant step-size in
the high-dimensional regime. We prove limit theorems for the trajectories of summary …
the high-dimensional regime. We prove limit theorems for the trajectories of summary …
Ecological communities with Lotka-Volterra dynamics
G Bunin - Physical Review E, 2017 - APS
Ecological communities in heterogeneous environments assemble through the combined
effect of species interaction and migration. Understanding the effect of these processes on …
effect of species interaction and migration. Understanding the effect of these processes on …
Sampling with flows, diffusion, and autoregressive neural networks from a spin-glass perspective
Recent years witnessed the development of powerful generative models based on flows,
diffusion, or autoregressive neural networks, achieving remarkable success in generating …
diffusion, or autoregressive neural networks, achieving remarkable success in generating …
Enhanced associative memory, classification, and learning with active dynamics
Motivated by advances in the field of active matter where nonequilibrium forcing has been
shown to activate new assembly pathways, here we study how nonequilibrium driving in …
shown to activate new assembly pathways, here we study how nonequilibrium driving in …
Violation of the fluctuation–dissipation theorem in glassy systems: basic notions and the numerical evidence
A Crisanti, F Ritort - Journal of Physics A: Mathematical and …, 2003 - iopscience.iop.org
This review reports on the research done during past years on violations of the fluctuation–
dissipation theorem (FDT) in glassy systems. It is focused on the existence of a quasi …
dissipation theorem (FDT) in glassy systems. It is focused on the existence of a quasi …
The landscape of the spiked tensor model
We consider the problem of estimating a large rank‐one tensor u⊗ k∈(ℝn)⊗ k, k≥ 3, in
Gaussian noise. Earlier work characterized a critical signal‐to‐noise ratio λ Bayes= O (1) …
Gaussian noise. Earlier work characterized a critical signal‐to‐noise ratio λ Bayes= O (1) …
The high-dimensional asymptotics of first order methods with random data
We study a class of deterministic flows in ${\mathbb R}^{d\times k} $, parametrized by a
random matrix ${\boldsymbol X}\in {\mathbb R}^{n\times d} $ with iid centered subgaussian …
random matrix ${\boldsymbol X}\in {\mathbb R}^{n\times d} $ with iid centered subgaussian …
Low-degree hardness of random optimization problems
D Gamarnik, A Jagannath… - 2020 IEEE 61st Annual …, 2020 - ieeexplore.ieee.org
We consider the problem of finding nearly optimal solutions of optimization problems with
random objective functions. Such problems arise widely in the theory of random graphs …
random objective functions. Such problems arise widely in the theory of random graphs …