YoVDO

Insights on Gradient-Based Algorithms in High-Dimensional Learning

Offered By: Simons Institute via YouTube

Tags

Machine Learning Courses Data Analysis Courses Statistical Physics Courses Stochastic Gradient Descent Courses Dynamical Mean Field Theory Courses Langevin Dynamics Courses

Course Description

Overview

Explore gradient-based algorithms in high-dimensional learning through this Richard M. Karp Distinguished Lecture. Delve into the analysis of gradient descent algorithms and their noisy variants in nonconvex settings. Examine several high-dimensional statistical learning problems where gradient-based algorithm performance can be analyzed precisely. Discover how statistical physics provides exact closed solutions for algorithm performance in the high-dimensional limit. Cover topics including the spiked mixed matrix-tensor model, perceptron, and phase retrieval. Gain insights into dynamical mean-field theory, Langevin dynamics, and stochastic gradient descent. Investigate phase diagrams, landscape analysis, and the teacher-student perceptron model. Understand the behavior of gradient descent in phase retrieval and explore theories for over-parameterized landscapes.

Syllabus

Intro
WORKHORSE OF MACHINE LEARNIN
IN DEEP LEARNING
STRATEGY
WHY THIS MODEL?
ESTIMATORS
GRADIENT-BASED ALGORITHMS
DYNAMICAL MEAN FIELD THEORY
LANGEVIN STATE EVOLUTION (NUMERICAL SOLUTION)
LANGEVIN PHASE DIAGRAM
GRADIENT-FLOW PHASE DIAGRAM
POPULAR "EXPLANATION"
SPURIOUS MINIMA DO NOT NECESSARILY CAUSE GF TO FAIL
WHAT IS GOING ON?
TRANSITION RECIPE
TRANSITION CONJECTUR
LANDSCAPE ANALYSIS
CONCLUSION ON SPIKED MATRIX-TENSOR MODEL
TEACHER-NEURAL SETTING
TEACHER STUDENT PERCEPTRON
PHASE RETRIEVAL: OPTIMAL SOLUTION
GRADIENT DESCENT FOR PHASE RETRIEVAI
PERFORMANCE OF GRADIENT DESCENT
GRADIENT DESCENT NUMERICALLY
TOWARDS A THEORY
OVER-PARAMETRISED LANDSPACE
STOCHASTIC GRADIENT DESCENT
DYNAMICAL MEAN-FIELD THEOR Mignaco, Urbani, Krzakala, LZ, 2006.06098
DMFT FOLLOWS THE WHOLE TRAJECTORY


Taught by

Simons Institute

Related Courses

Estimating Normalizing Constants for Log-Concave Distributions - Algorithms and Lower Bounds
Association for Computing Machinery (ACM) via YouTube
Sampling Using Diffusion Processes, From Langevin to Schrödinger
Simons Institute via YouTube
Stochastic Analysis and Geometric Functional Inequalities
Hausdorff Center for Mathematics via YouTube
Speeding up Langevin Dynamics by Mixing
Society for Industrial and Applied Mathematics via YouTube
Optimal Transport and High-Dimensional Probability - Langevin Dynamics via Calculus
Institute for Mathematical Sciences via YouTube