Zobrazeno 1 - 10
of 856
pro vyhledávání: '"Serrurier, A."'
Autor:
Bethune, Louis, Massena, Thomas, Boissin, Thibaut, Prudent, Yannick, Friedrich, Corentin, Mamalet, Franck, Bellet, Aurelien, Serrurier, Mathieu, Vigouroux, David
State-of-the-art approaches for training Differentially Private (DP) Deep Neural Networks (DNN) face difficulties to estimate tight bounds on the sensitivity of the network's layers, and instead rely on a process of per-sample gradient clipping. This
Externí odkaz:
http://arxiv.org/abs/2305.16202
Autor:
Bethune, Louis, Novello, Paul, Boissin, Thibaut, Coiffier, Guillaume, Serrurier, Mathieu, Vincenot, Quentin, Troya-Galvis, Andres
We propose a new method, dubbed One Class Signed Distance Function (OCSDF), to perform One Class Classification (OCC) by provably learning the Signed Distance Function (SDF) to the boundary of the support of any distribution. The distance to the supp
Externí odkaz:
http://arxiv.org/abs/2303.01978
Publikováno v:
Conference on Neural Information Processing Systems (NeurIPS), Neural Information Processing Systems Foundation, Dec 2023, New Orleans (Louisiana), United States
Input gradients have a pivotal role in a variety of applications, including adversarial attack algorithms for evaluating model robustness, explainable AI techniques for generating Saliency Maps, and counterfactual explanations.However, Saliency Maps
Externí odkaz:
http://arxiv.org/abs/2206.06854
Audio-based classification techniques on body sounds have long been studied to aid in the diagnosis of respiratory diseases. While most research is centered on the use of cough as the main biomarker, other body sounds also have the potential to detec
Externí odkaz:
http://arxiv.org/abs/2204.10581
Autor:
Bakkay, Mohamed Chafik, Serrurier, Mathieu, Burda, Valentin Kivachuk, Dupuy, Florian, Cabrera-Gutierrez, Naty Citlali, Zamo, Michael, Mader, Maud-Alix, Mestre, Olivier, Oller, Guillaume, Jouhaud, Jean-Christophe, Terray, Laurent
Precipitation nowcasting is of great importance for weather forecast users, for activities ranging from outdoor activities and sports competitions to airport traffic management. In contrast to long-term precipitation forecasts which are traditionally
Externí odkaz:
http://arxiv.org/abs/2203.13263
Publikováno v:
Sensors, Vol 24, Iss 19, p 6176 (2024)
Audio-based classification techniques for body sounds have long been studied to aid in the diagnosis of respiratory diseases. While most research is centered on the use of coughs as the main acoustic biomarker, other body sounds also have the potenti
Externí odkaz:
https://doaj.org/article/70fbc1d4dfa04866ad67289ea914210f
Autor:
Baaj, Ismaïl, Bouraoui, Zied, Cornuéjols, Antoine, Denœux, Thierry, Destercke, Sébastien, Dubois, Didier, Lesot, Marie-Jeanne, Marques-Silva, João, Mengin, Jérôme, Prade, Henri, Schockaert, Steven, Serrurier, Mathieu, Strauss, Olivier, Vrain, Christel
Publikováno v:
In International Journal of Approximate Reasoning August 2024 171
Autor:
Béthune, Louis, Boissin, Thibaut, Serrurier, Mathieu, Mamalet, Franck, Friedrich, Corentin, González-Sanz, Alberto
Lipschitz constrained networks have gathered considerable attention in the deep learning community, with usages ranging from Wasserstein distance estimation to the training of certifiably robust classifiers. However they remain commonly considered as
Externí odkaz:
http://arxiv.org/abs/2104.05097
We present a new algorithm to solve min-max or min-min problems out of the convex world. We use rigidity assumptions, ubiquitous in learning, making our method applicable to many optimization problems. Our approach takes advantage of hidden regularit
Externí odkaz:
http://arxiv.org/abs/2007.08810
Autor:
Dupuy, Florian, Mestre, Olivier, Serrurier, Mathieu, Bakkay, Mohamed Chafik, Burdá, Valentin Kivachuk, Cabrera-Gutiérrez, Naty Citlali, Jouhaud, Jean-Christophe, Mader, Maud-Alix, Oller, Guillaume, Zamo, Michaël
Cloud cover is crucial information for many applications such as planning land observation missions from space. It remains however a challenging variable to forecast, and Numerical Weather Prediction (NWP) models suffer from significant biases, hence
Externí odkaz:
http://arxiv.org/abs/2006.16678