Pascal Germain

Reconstruction functions are pivotal in sample compression theory, a framework for deriving tight generalization bounds. From a small sample… (see more) of the training set (the compression set) and an optional stream of information (the message), they recover a predictor previously learned from the whole training set. While usually fixed, we propose to learn reconstruction functions. To facilitate the optimization and increase the expressiveness of the message, we derive a new sample compression generalization bound for real-valued messages. From this theoretical analysis, we then present a new hypernetwork architecture that outputs predictors with tight generalization guarantees when trained using an original meta-learning framework. The results of promising preliminary experiments are then reported.

2024-10-09

NeurIPS.cc/2024/Workshop/Compression (published)

Sample compression unleashed : New generalization bounds for real valued losses

Mathieu Bazinet

Valentina Zantedeschi

2024-10-09

NeurIPS.cc/2024/Workshop/Compression (published)

Seeking Interpretability and Explainability in Binary Activated Neural Networks

Benjamin Leblanc

2024-07-10

Communications in Computer and Information Science (published)

Phoneme Discretized Saliency Maps for Explainable Detection of AI-Generated Voice

Shubham Gupta

Mirco Ravanelli

Cem Subakan

2024-06-14

ArXiv (preprint)

A general framework for the practical disintegration of PAC-Bayesian bounds

Paul Viallard

Amaury Habrard

Emilie Morvant

2023-10-11

Machine-mediated learning (published)

Statistical Guarantees for Variational Autoencoders using PAC-Bayesian Theory

Sokhna Diarra Mbacke

Florence Clerc

Statistical Guarantees for Variational Autoencoders using PAC-Bayesian Theory

Sokhna Diarra Mbacke

Florence Clerc

2023-09-21

NeurIPS.cc/2023/Conference (spotlight)

PAC-Bayesian Generalization Bounds for Adversarial Generative Models

Sokhna Diarra Mbacke

Florence Clerc

2023-07-03

Proceedings of the 40th International Conference on Machine Learning (published)

Invariant Causal Set Covering Machines

Thibaud Godon

Baptiste Bauvin

Jacques Corbeil

Alexandre Drouin

2023-06-07

ArXiv (preprint)

PAC-Bayesian Learning of Aggregated Binary Activated Neural Networks with Probabilities over Representations

Louis Fortier-Dubois

Gaël Letarte

Benjamin Leblanc

François Laviolette

2023-06-05

Proceedings of the Canadian Conference on Artificial Intelligence (published)

PAC-Bayesian Generalization Bounds for Adversarial Generative Models

Sokhna Diarra Mbacke

Florence Clerc

We extend PAC-Bayesian theory to generative models and develop generalization bounds for models based on the Wasserstein distance and the to… (see more)tal variation distance. Our first result on the Wasserstein distance assumes the instance space is bounded, while our second result takes advantage of dimensionality reduction. Our results naturally apply to Wasserstein GANs and Energy-Based GANs, and our bounds provide new training objectives for these two. Although our work is mainly theoretical, we perform numerical experiments showing non-vacuous generalization bounds for Wasserstein GANs on synthetic datasets.

2023-01-01

ICML (published)

Sample Boosting Algorithm (SamBA) - An interpretable greedy ensemble classifier based on local expertise for fat data

Baptiste Bauvin

Cécile Capponi

Florence Clerc

Sokol Koço

Jacques Corbeil

2023-01-01

UAI (published)

proceedings.mlr.press