no code implementations • 15 Dec 2022 • Clara Carlier, Arnaud Franju, Matthieu Lerasle, Mathias Obrebski
Recent developments of advanced driver-assistance systems necessitate an increasing number of tests to validate new technologies.
no code implementations • 12 Mar 2020 • Geoffrey Chinot, Matthieu Lerasle
For low signal to noise ratio, we also provide lower bound holding with large probability.
no code implementations • 11 Sep 2019 • Guillaume Maillard, Sylvain Arlot, Matthieu Lerasle
Aggregated hold-out (Agghoo) is a method which averages learning rules selected by hold-out (that is, cross-validation with a single split).
no code implementations • 28 Aug 2019 • Matthieu Lerasle
These notes gather recent results on robust statistical learning theory.
no code implementations • 17 May 2019 • Christophe Giraud, Yann Issartel, Luc Lehéricy, Matthieu Lerasle
This paper shows that sublinear regret is achievable in the case where the graph is generated according to a Stochastic Block Model (SBM) with two communities.
no code implementations • 13 Feb 2018 • Matthieu Lerasle, Zoltan Szabo, Timothee Mathieu, Guillaume Lecue
Mean embeddings provide an extremely flexible and powerful tool in machine learning and statistics to represent probability distributions and define a semi-metric (MMD, maximum mean discrepancy; also called N-distance or energy distance), with numerous successful applications.
no code implementations • 22 Oct 2012 • Sylvain Arlot, Matthieu Lerasle
Then, we compute the variance of V-fold cross-validation and related criteria, as well as the variance of key quantities for model selection performance.