Mediaspace scheduled maintenance: Aug 25, 2026 07:00 - 12:00 AM. During this time, videos will be temporarily unavailable. Check status updates.

Course

PHYS-642: Statistical physics for optimization & learning

Lectures in this course (44)

Covers the Curie-Weiss model in Statistical Physics, including magnetization probability, free entropy, and the cavity method.

Explores properties of clusters and colorability threshold in graph coloring, including average connectivity and rigidity.

Explores Q(m) as a population behavior indicator and its implications.

Explores Belief Propagation for graph coloring and its convergence properties.

Explores Replica Symmetry Breaking in the Random Energy Model, discussing configurational entropy and condensation.

Explores Replica Symmetry Breaking in the Random Energy Model, focusing on entropy and probability states.

Explores graphical models, factor graphs, and probabilistic inferences in complex systems.

Explores a variety of open problems in graph theory and computational complexity, challenging students to analyze and solve complex issues.

Covers inference problems related to the Spin Glass Game and the challenges of making mistakes with preb P.

Covers the basics of spin glasses and the mathematical equations used to model them.

Explores state evolution, Gaussian iteration, and convergence criteria in iterative algorithms.

Explores the Spike Wigner model, Bayesian denoising, state evolution, and spectral methods in matrix analysis.

Covers the Random Constraint Satisfaction Problem and explores methods for solving constraint satisfaction problems.

Covers the Stochastic Block Model and its application in community detection, exploring its mathematical formulation and challenges.

Covers a proof technique for the spiked model using matrix-based calculations and state evolution.

Explores Generalized Linear Models, Bayesian methods, compressed sensing, and perception in high-dimensional statistics.

Explores advanced graph coloring concepts, including planted coloring, rigidity threshold, and frozen variables in BP fixed points.

Covers the GAMP algorithm for iterative signal reconstruction and introduces proximal gradient descent for L1 minimization problems.

Delves into applying the GAMP algorithm to simplify the lasso problem and analyze optimization challenges in neural networks.

Offers insights into the statistical physics of learning, exploring the relationship between neural network structure and disordered systems.