6 papers
Are Current Continual Learning Methods Truly Agnostic? Introducing OPRE, a Step Toward Agnostic Continual Learning
Raphaël Bayle, Martial Mermillod, Robert M. French
In order to achieve Continual Learning (CL), the problem of catastrophic forgetting, one that has plagued neural networks since their inception, must be overcome. The evaluation of…
How Foundational are Foundation Models for Time Series Forecasting?
Nouha Karaouli, Denis Coquenet, Elisa Fromont +2
Foundation Models are designed to serve as versatile embedding machines, with strong zero shot capabilities and superior generalization performance when fine-tuned on diverse downs…
Foundation vs. Specialized Models: Evaluating Catastrophic Forgetting in Continual Time Series Forecasting
Nouha Karaouli, Denis Coquenet, Elisa Fromont +2
While Time Series Foundation Models (TSFMs) excel in zero-shot tasks, their behavior under continual fine tuning is poorly understood. We present the first systematic study of cata…
Graph-based methods coupled with specific distributional distances for adversarial attack detection
Dwight Nwaigwe, Lucrezia Carboni, Martial Mermillod +2
Artificial neural networks are prone to being fooled by carefully perturbed inputs which cause an egregious misclassification. These \textit{adversarial} attacks have been the focu…
Impact of Spatial Frequency Based Constraints on Adversarial Robustness
Rémi Bernhard, Pierre-Alain Moellic, Martial Mermillod +4
Adversarial examples mainly exploit changes to input pixels to which humans are not sensitive to, and arise from the fact that models make decisions based on uninterpretable featur…
Modelling risk-taking behaviour of avalanche accident victims
Robin Couret, Carole Adam, Martial Mermillod
Each year, over 15000 requests for mountain rescue are counted in France. Avalanche accidents represent 39\% of reports, and are therefore our focus in this study. Modelling the be…