no code implementations • 7 Feb 2024 • Camila Fernandez, Pierre Gaillard, Joseph de Vilmarest, Olivier Wintenberger
We introduce an online mathematical framework for survival analysis, allowing real time adaptation to dynamic environments and censored data.
no code implementations • 24 Jan 2023 • Joseph de Vilmarest, Jethro Browell, Matteo Fasiolo, Yannig Goude, Olivier Wintenberger
The proliferation of local generation, demand response, and electrification of heat and transport are changing the fundamental drivers of electricity load and increasing the complexity of load modelling and forecasting.
no code implementations • 18 Jan 2023 • Maxime Haddouche, Olivier Wintenberger, Benjamin Guedj
Optimistic Online Learning algorithms have been developed to exploit expert advices, assumed optimistically to be always useful.
no code implementations • 25 May 2022 • Antoine Godichon-Baggioni, Nicklas Werge, Olivier Wintenberger
This paper addresses stochastic optimization in a streaming setting with time-dependent and biased gradient estimates.
no code implementations • 15 Sep 2021 • Antoine Godichon-Baggioni, Nicklas Werge, Olivier Wintenberger
We provide non-asymptotic convergence rates of various gradient-based algorithms; this includes the famous Stochastic Gradient (SG) descent (a. k. a.
no code implementations • 16 Apr 2021 • Joseph de Vilmarest, Olivier Wintenberger
We introduce an augmented model in which the variances are represented as auxiliary gaussian latent variables in a tracking mode.
no code implementations • 12 Feb 2021 • Clément Dombry, Charles Tillier, Olivier Wintenberger
We consider regular variation for marked point processes with independent heavy-tailed marks and prove a single large point heuristic: the limit measure is concentrated on the cone of point measures with one single point.
Point Processes Probability
no code implementations • 1 Feb 2021 • Olivier Wintenberger
We introduce a general framework of stochastic online convex optimization to obtain fast-rate stochastic regret bounds.
1 code implementation • 3 Jun 2020 • Nicklas Werge, Olivier Wintenberger
An investigation of the convergence properties of the QML procedure in a general conditionally heteroscedastic time series model is conducted, and the classical batch optimization routines extended to the framework of streaming and large-scale problems.
no code implementations • 26 Feb 2020 • Eric Adjakossa, Yannig Goude, Olivier Wintenberger
In this article, we aim at improving the prediction of expert aggregation by using the underlying properties of the models that provide expert predictions.
no code implementations • 10 Feb 2020 • Joseph de Vilmarest, Olivier Wintenberger
Second, for generalized linear regressions, we provide a martingale analysis of the excess risk in the local phase, improving existing ones in bounded stochastic optimization.
1 code implementation • 4 Jul 2019 • Vincent Margot, Jean-Patrick Baudry, Frédéric Guilloux, Olivier Wintenberger
The proof of the consistency is based on a control of the error of the empirical estimation of conditional expectations which is interesting on its own.
no code implementations • 1 Jul 2019 • Meyer Nicolas, Olivier Wintenberger
Regular variation provides a convenient theoretical framework to study large events.
no code implementations • 26 Feb 2019 • Joseph De Vilmarest, Olivier Wintenberger
We consider online optimization procedures in the context of logistic regression, focusing on the Extended Kalman Filter (EKF).
2 code implementations • 12 Jul 2018 • Vincent Margot, Jean-Patrick Baudry, Frederic Guilloux, Olivier Wintenberger
RIPE is a novel deterministic and easily understandable prediction algorithm developed for continuous and discrete ordered data.
no code implementations • NeurIPS 2018 • Pierre Gaillard, Olivier Wintenberger
setting, we establish new risk bounds that are adaptive to the sparsity of the problem and to the regularity of the risk (ranging from a rate 1 / $\sqrt T$ for general convex risk to 1 /T for strongly convex risk).
no code implementations • 19 Aug 2016 • Niklas Thiemann, Christian Igel, Olivier Wintenberger, Yevgeny Seldin
We propose a new PAC-Bayesian bound and a way of constructing a hypothesis space, so that the bound is convex in the posterior distribution and also convex in a trade-off parameter between empirical performance of the posterior distribution and its complexity.
no code implementations • 4 Apr 2014 • Olivier Wintenberger
The exponential weights include an accuracy term and a second order term that is a proxy of the quadratic variation as in Hazan and Kale (2010).