Unit

Theory of Machine Learning Laboratory

Laboratory

Summary

The Theory Laboratory in Automatic Learning (TML) at EPFL focuses on developing algorithmic and theoretical tools to enhance the understanding and robustness of machine learning. Their research spans topics such as adversarial machine learning, deep learning theory, stochastic optimization, and MCMC algorithms. The lab's team is actively involved in investigating various aspects of machine learning, including model agnostic methods, norm regularization, and sharpness-aware minimization. With a strong emphasis on advancing the field of machine learning, TML is dedicated to exploring innovative approaches to improve the performance and reliability of machine learning models.

Deep Learning Theory Through the Lens of Diagonal Linear Networks

Scott William Pesme

In this PhD manuscript, we explore optimisation phenomena which occur in complex neural networks through the lens of

2

-layer diagonal linear networks. This rudimentary architecture, which consists of a two layer feedforward linear network with a diagonal ...

EPFL2024

Theory of Machine Learning Laboratory

Graph Chatbot

Deep Learning Theory Through the Lens of Diagonal Linear Networks

Deep Learning Generalization with Limited and Noisy Labels

Fundamental Limits in Statistical Learning Problems: Block Models and Neural Networks

Deep Learning Theory Through the Lens of Diagonal Linear Networks

Fundamental Limits in Statistical Learning Problems: Block Models and Neural Networks

Deep Learning Generalization with Limited and Noisy Labels