Publications associées à Phase diagram of Stochastic Gradient Descent in high-dimensional two-layer neural networks

Deep Learning Theory Through the Lens of Diagonal Linear Networks

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

Topics in statistical physics of high-dimensional machine learning

Hugo Chao Cui

In the past few years, Machine Learning (ML) techniques have ushered in a paradigm shift, allowing the harnessing of ever more abundant sources of data to automate complex tasks. The technical workhorse behind these important breakthroughs arguably lies in ...

EPFL2024

Understanding generalization and robustness in modern deep learning

Maksym Andriushchenko

In this thesis, we study two closely related directions: robustness and generalization in modern deep learning. Deep learning models based on empirical risk minimization are known to be often non-robust to small, worst-case perturbations known as adversari ...

EPFL2024

Leveraging Continuous Time to Understand Momentum When Training Diagonal Linear Networks

Nicolas Henri Bernard Flammarion, Hristo Georgiev Papazov, Scott William Pesme

In this work, we investigate the effect of momentum on the optimisation trajectory of gradient descent. We leverage a continuous-time approach in the analysis of momentum gradient descent with step size

\gamma

and momentum parameter

\beta

that allows u ...

2024

On the Generalization of Stochastic Gradient Descent with Momentum

Volkan Cevher, Kimon Antonakopoulos

While momentum-based accelerated variants of stochastic gradient descent (SGD) are widely used when training machine learning models, there is little theoretical understanding on the generalization error of such methods. In this work, we first show that th ...

Brookline2024

Explainable Face Verification via Feature-Guided Gradient Backpropagation

Touradj Ebrahimi, Yuhang Lu, Zewei Xu

Recent years have witnessed significant advance- ment in face recognition (FR) techniques, with their applications widely spread in people’s lives and security-sensitive areas. There is a growing need for reliable interpretations of decisions of such syste ...

2024

Revisiting Character-level Adversarial Attacks for Language Models

Volkan Cevher, Grigorios Chrysos, Fanghui Liu, Yongtao Wu, Elias Abad Rocamora

Adversarial attacks in Natural Language Processing apply perturbations in the character or token levels. Token-level attacks, gaining prominence for their use of gradient-based methods, are susceptible to altering sentence semantics, leading to invalid adv ...

2024

Deep Learning Generalization with Limited and Noisy Labels

Mahsa Forouzesh

Deep neural networks have become ubiquitous in today's technological landscape, finding their way in a vast array of applications. Deep supervised learning, which relies on large labeled datasets, has been particularly successful in areas such as image cla ...

EPFL2023

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

Elisabetta Cornacchia

This thesis focuses on two selected learning problems: 1) statistical inference on graphs models, and, 2) gradient descent on neural networks, with the common objective of defining and analysing the measures that characterize the fundamental limits.In the ...

EPFL2023

Hamiltonian Deep Neural Networks Guaranteeing Non-Vanishing Gradients by Design

Giancarlo Ferrari Trecate, Luca Furieri, Clara Lucía Galimberti, Liang Xu

Deep Neural Networks (DNNs) training can be difficult due to vanishing and exploding gradients during weight optimization through backpropagation. To address this problem, we propose a general class of Hamiltonian DNNs (H-DNNs) that stem from the discretiz ...

2023

Phase diagram of Stochastic Gradient Descent in high-dimensional two-layer neural networks

Graph Chatbot

Chattez avec Graph Search

Deep Learning Theory Through the Lens of Diagonal Linear Networks

Topics in statistical physics of high-dimensional machine learning

Understanding generalization and robustness in modern deep learning

Leveraging Continuous Time to Understand Momentum When Training Diagonal Linear Networks

On the Generalization of Stochastic Gradient Descent with Momentum

Explainable Face Verification via Feature-Guided Gradient Backpropagation

Revisiting Character-level Adversarial Attacks for Language Models

Deep Learning Generalization with Limited and Noisy Labels

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

Hamiltonian Deep Neural Networks Guaranteeing Non-Vanishing Gradients by Design

Topics in statistical physics of high-dimensional machine learning

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

Deep Learning Generalization with Limited and Noisy Labels

Deep Learning Theory Through the Lens of Diagonal Linear Networks

Understanding generalization and robustness in modern deep learning

Revisiting Character-level Adversarial Attacks for Language Models

Explainable Face Verification via Feature-Guided Gradient Backpropagation

Hamiltonian Deep Neural Networks Guaranteeing Non-Vanishing Gradients by Design

Leveraging Continuous Time to Understand Momentum When Training Diagonal Linear Networks

On the Generalization of Stochastic Gradient Descent with Momentum