Combinatorial Invariance: A Case Study of Pure Math / Machine Learning Interaction - Geordie Williamson
Institute for Advanced Study via YouTube
Overview
This course explores the intersection of pure mathematics and machine learning through the study of the combinatorial invariance conjecture. The learning outcomes include understanding the application of modern machine learning techniques in approaching problems in pure mathematics, exploring how machine learning models work, and extracting new mathematics from these models. The course teaches skills such as working with neural nets, convolutional nets, training machine learning models, and visualizing combinatorial invariance. The teaching method involves lectures, examples, and discussions on predicting KLL polynomials and analyzing the saliency of the results. The intended audience for this course includes individuals interested in the application of machine learning in pure mathematics and those looking to explore the connection between theoretical concepts and practical computational methods.
Syllabus
Introduction
Motivation
Perceptron
Neural nets
Convolutional nets
convolutional neural nets
geometries
generalizations
training
machine learning for mathematicians
bruja graph
analytic polynomials
Examples
Visualizing combinatorial invariance
Predicting KLL polynomials
Saliency
Taught by
Institute for Advanced Study