Publications associées à Laboratoire de systèmes d'information et d'inférence

Robustness in deep learning: The good (width), the bad (depth), and the ugly (initialization)

Volkan Cevher, Grigorios Chrysos, Fanghui Liu, Zhenyu Zhu

We study the average robustness notion in deep neural networks in (selected) wide and narrow, deep and shallow, as well as lazy and non-lazy training settings. We prove that in the under-parameterized setting, width has a negative effect while it improves ...

2022

A Structured Dictionary Perspective on Implicit Neural Representations

Pascal Frossard, Guillermo Ortiz Jimenez, Gizem Yüce, Beril Besbinar

Implicit neural representations (INRs) have recently emerged as a promising alternative to classical discretized representations of signals. Nevertheless, despite their practical success, we still do not understand how INRs represent signals. We propose a ...

IEEE COMPUTER SOC2022

Decentralized Semi-supervised Learning over Multitask Graphs

Ali H. Sayed, Roula Nassif, Elsa Rizk

In network semi-supervised learning problems, only a subset of the network nodes is able to access the data labeling. This paper formulates a decentralized optimization problem where agents have individual decision rules to estimate, subject to the conditi ...

IEEE2022

On Breathing Pattern Information in Synthetic Speech

Mathew Magimai Doss, Zohreh Mostaani

The respiratory system is an integral part of human speech production. As a consequence, there is a close relation between respiration and speech signal, and the produced speech signal carries breathing pattern related information. Speech can also be gener ...

ISCA-INT SPEECH COMMUNICATION ASSOC2022

Learning to Align Sequential Actions in the Wild

Pascal Fua, Bugra Tekin, Weizhe Liu

State-of-the-art methods for self-supervised sequential action alignment rely on deep networks that find correspon- dences across videos in time. They either learn frame-to- frame mapping across sequences, which does not leverage temporal information, or a ...

IEEE2022

System and method for privacy-preserving distributed training of neural network models on distributed datasets

Jean-Pierre Hubaux, Juan Ramón Troncoso-Pastoriza, Jean-Philippe Léonard Bossuat, Apostolos Pyrgelis, David Jules Froelicher, Sinem Sav

A computer-implemented method and a distributed computer system (100) for privacy- preserving distributed training of a global neural network model on distributed datasets (DS1 to DSn). The system has a plurality of data providers (DP1 to DPn) being commun ...

2022

From Undercomplete to Sparse Overcomplete Autoencoders to Improve LF-MMI Speech Recognition

Hervé Bourlard, Selen Hande Kabil

Starting from a strong Lattice-Free Maximum Mutual Information (LF-MMI) baseline system, we explore different autoencoder configurations to enhance Mel-Frequency Cepstral Coefficients (MFCC) features. Autoencoders are expected to generate new MFCC features ...

ISCA-INT SPEECH COMMUNICATION ASSOC2022

Towards Robust Vision Transformer

Shaokai Ye, Yuan He

Recent advances on Vision Transformer (ViT) and its improved variants have shown that self-attention-based networks surpass traditional Convolutional Neural Networks (CNNs) in most vision tasks. However, existing ViTs focus on the standard accuracy and com ...

IEEE COMPUTER SOC2022

No-regret learning in games with noisy feedback: Faster rates and adaptivity via learning rate separation

Volkan Cevher, Kimon Antonakopoulos, Ya-Ping Hsieh

We examine the problem of regret minimization when the learner is involved in a continuous game with other optimizing agents: in this case, if all players follow a no-regret algorithm, it is possible to achieve significantly lower regret relative to fully ...

2022

Chattez avec Graph Search

Posez n’importe quelle question sur les cours, conférences, exercices, recherches, actualités, etc. de l’EPFL ou essayez les exemples de questions ci-dessous.

AVERTISSEMENT : Le chatbot Graph n'est pas programmé pour fournir des réponses explicites ou catégoriques à vos questions. Il transforme plutôt vos questions en demandes API qui sont distribuées aux différents services informatiques officiellement administrés par l'EPFL. Son but est uniquement de collecter et de recommander des références pertinentes à des contenus que vous pouvez explorer pour vous aider à répondre à vos questions.

Laboratoire de systèmes d'information et d'inférence

Robustness in deep learning: The good (width), the bad (depth), and the ugly (initialization)

A Structured Dictionary Perspective on Implicit Neural Representations

Decentralized Semi-supervised Learning over Multitask Graphs

On Breathing Pattern Information in Synthetic Speech

Learning to Align Sequential Actions in the Wild

System and method for privacy-preserving distributed training of neural network models on distributed datasets

From Undercomplete to Sparse Overcomplete Autoencoders to Improve LF-MMI Speech Recognition

Towards Robust Vision Transformer

Annotation-efficient image anomaly detection

No-regret learning in games with noisy feedback: Faster rates and adaptivity via learning rate separation

Graph Chatbot

Chattez avec Graph Search

Robustness in deep learning: The good (width), the bad (depth), and the ugly (initialization)

No-regret learning in games with noisy feedback: Faster rates and adaptivity via learning rate separation

Annotation-efficient image anomaly detection

A Structured Dictionary Perspective on Implicit Neural Representations

Decentralized Semi-supervised Learning over Multitask Graphs

Learning to Align Sequential Actions in the Wild

System and method for privacy-preserving distributed training of neural network models on distributed datasets

From Undercomplete to Sparse Overcomplete Autoencoders to Improve LF-MMI Speech Recognition

On Breathing Pattern Information in Synthetic Speech

Towards Robust Vision Transformer