A Study of the Mathematics of Deep Learning

"Deep Learning"/"Deep Neural Nets" is a technological marvel that is now increasingly deployed at the cutting-edge of artificial intelligence tasks. This dramatic success of deep learning in the last few years has been hinged on an enormous amount of heuristics and it has turned...

Full description

Saved in:
Bibliographic Details
Published in:arXiv.org 2021-04
Main Author: Mukherjee, Anirbit
Format: Article
Language:English
Subjects:
Adaptive algorithms
Algorithms
Applications of mathematics
Artificial intelligence
Circuits
Complexity
Deep learning
Function space
Machine learning
Mathematical analysis
Neural networks
Theorems
Online Access:Get full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:"Deep Learning"/"Deep Neural Nets" is a technological marvel that is now increasingly deployed at the cutting-edge of artificial intelligence tasks. This dramatic success of deep learning in the last few years has been hinged on an enormous amount of heuristics and it has turned out to be a serious mathematical challenge to be able to rigorously explain them. In this thesis, submitted to the Department of Applied Mathematics and Statistics, Johns Hopkins University we take several steps towards building strong theoretical foundations for these new paradigms of deep-learning. In chapter 2 we show new circuit complexity theorems for deep neural functions and prove classification theorems about these function spaces which in turn lead to exact algorithms for empirical risk minimization for depth 2 ReLU nets. We also motivate a measure of complexity of neural functions to constructively establish the existence of high-complexity neural functions. In chapter 3 we give the first algorithm which can train a ReLU gate in the realizable setting in linear time in an almost distribution free set up. In chapter 4 we give rigorous proofs towards explaining the phenomenon of autoencoders being able to do sparse-coding. In chapter 5 we give the first-of-its-kind proofs of convergence for stochastic and deterministic versions of the widely used adaptive gradient deep-learning algorithms, RMSProp and ADAM. This chapter also includes a detailed empirical study on autoencoders of the hyper-parameter values at which modern algorithms have a significant advantage over classical acceleration based methods. In the last chapter 6 we give new and improved PAC-Bayesian bounds for the risk of stochastic neural nets. This chapter also includes an experimental investigation revealing new geometric properties of the paths in weight space that are traced out by the net during the training.
ISSN:2331-8422