Výsledky vyhledávání - "Alessandro, Lazaric"

Autor: Rui Yuan, Alessandro Lazaric, Robert M. Gower

We propose a new globally convergent stochastic second order method. Our starting point is the development of a new Sketched Newton-Raphson (SNR) method for solving large scale nonlinear equations of the form $F(x)=0$ with $F:\mathbb{R}^p \rightarrow

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::9e7e608f4184eaafffd252babef283f3

Zobrazit plný text záznamu

Improved Algorithms for Conservative Exploration in Bandits

Autor: Alessandro Lazaric, Evrard Garcelon, Mohammad Ghavamzadeh, Matteo Pirotta

Publikováno v: AAAI

In many fields such as digital marketing, healthcare, finance, and robotics, it is common to have a well-tested and reliable baseline policy running in production (e.g., a recommender system). Nonetheless, the baseline policy is often suboptimal. In

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::41fdf18f1530f1c9aa5ec47de20244e6

Zobrazit plný text záznamu

An Asymptotically Optimal Primal-Dual Incremental Algorithm for Contextual Linear Bandits

Autor: Tirinzoni, Andrea, Matteo, Pirotta, Restelli, Marcello, Alessandro, Lazaric

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=od______3731::cd52cc395f4e9ab293dd3a8f058e5ab9
http://hdl.handle.net/11311/1167116

Zobrazit plný text záznamu

Word-order biases in deep-agent emergent communication

Autor: Eugene Kharitonov, Rahma Chaabouni, Marco Baroni, Emmanuel Dupoux, Alessandro Lazaric

Publikováno v: ACL 2019-57th Annual Meeting of the Association for Computational Linguistics
ACL 2019-57th Annual Meeting of the Association for Computational Linguistics, Jul 2019, Florence, Italy
ACL (1)
HAL

Sequence-processing neural networks led to remarkable progress on many NLP tasks. As a consequence, there has been increasing interest in understanding to what extent they process language as humans do. We aim here to uncover which biases such models

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::40f59af8b4cb632aa908dc7ebdbfc3bd
https://hal.archives-ouvertes.fr/hal-02274157/file/1905.12330.pdf

Zobrazit plný text záznamu

Linear Thompson Sampling Revisited

Autor: Marc Abeille, Alessandro Lazaric

Publikováno v: AISTATS 2017-20th International Conference on Artificial Intelligence and Statistics
AISTATS 2017-20th International Conference on Artificial Intelligence and Statistics, Apr 2017, Fort Lauderdale, United States
Electron. J. Statist. 11, no. 2 (2017), 5165-5197

International audience; We derive an alternative proof for the regret of Thompson sampling (\ts) in the stochastic linear bandit setting. While we obtain a regret bound of order $\wt{O}(d^{3/2}\sqrt{T})$ as in previous results, the proof sheds new li

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::87799a182381b971fab8c58fce2f0248
https://hal.inria.fr/hal-01493561

Zobrazit plný text záznamu

LQG for Portfolio Optimization

Autor: Alessandro Lazaric, Xavier Brokmann, Marc Abeille, Emmanuel Sérié

Publikováno v: SSRN Electronic Journal.

We introduce a generic solver for dynamic portfolio allocation problems when the market exhibits return predictability, price impact and partial observability. We assume that the price modeling can be encoded into a linear state-space and we demonstr

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::914f848dfe66a7c31e6d6c31cf7cf7c1
https://doi.org/10.2139/ssrn.2863925

Zobrazit plný text záznamu

Sparse Multi-task Reinforcement Learning

Autor: Marcello Restelli, Alessandro Lazaric, Daniele Calandriello

Publikováno v: NIPS-Advances in Neural Information Processing Systems 26
NIPS-Advances in Neural Information Processing Systems 26, Dec 2014, Montreal, Canada
Scopus-Elsevier

International audience; In multi-task reinforcement learning (MTRL), the objective is to simultaneously learn multiple tasks and exploit their similarity to improve the performance w.r.t.\ single-task learning. In this paper we investigate the case w

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::bb412149ff1ffdc6c43b04a9534c563a
https://hal.inria.fr/hal-01073513/document

Zobrazit plný text záznamu

Truthful Learning Mechanisms for Multi-Slot Sponsored Search Auctions with Externalities

Autor: Nicola Gatti, Alessandro Lazaric, Marco Rocco, Francesco Trovò

Publikováno v: Artificial Intelligence
Artificial Intelligence, 2015, 227, pp.93-139. ⟨10.1016/j.artint.2015.05.012⟩
Artificial Intelligence, Elsevier, 2015, 227, pp.93-139. ⟨10.1016/j.artint.2015.05.012⟩

International audience; Sponsored Search Auctions (SSAs) constitute one of the most successful applications of microeconomic mechanisms. In mechanism design, auctions are usually designed to incentivize advertisers to bid their truthful valuations an

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::bc3ca4fc3fe5366665da4b558c3754c5
http://arxiv.org/abs/1405.2484

Zobrazit plný text záznamu

Regret Bounds for Reinforcement Learning with Policy Advice

Autor: Mohammad Gheshlaghi Azar, Emma Brunskill, Alessandro Lazaric

Publikováno v: ECML/PKDD-European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases
ECML/PKDD-European Conference on Machine Learning and Principles and Practice of Knowledge Discovery in Databases, Sep 2013, Prague, Czech Republic
Advanced Information Systems Engineering ISBN: 9783642387081
ECML/PKDD (1)

International audience; In some reinforcement learning problems an agent may be provided with a set of input policies, perhaps learned from prior experience or provided by advisors. We present a reinforcement learning with policy advice (RLPA) algori

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::c5bed6cb28c2f4542695c81d5d5cfbf7
http://arxiv.org/abs/1305.1027

Zobrazit plný text záznamu

Learning with stochastic inputs and adversarial outputs

Autor: Alessandro Lazaric, Rémi Munos

Publikováno v: Journal of Computer and System Sciences
Journal of Computer and System Sciences, Elsevier, 2012, 78 (5), pp.1516-1537. ⟨10.1016/j.jcss.2011.12.027⟩
Journal of Computer and System Sciences, 2012, 78 (5), pp.1516-1537. ⟨10.1016/j.jcss.2011.12.027⟩

International audience; Most of the research in online learning is focused either on the problem of adversarial classification (i.e., both inputs and labels are arbitrarily chosen by an adversary) or on the traditional supervised learning problem in

Externí odkaz: https://explore.openaire.eu/search/publication?articleId=doi_dedup___::8883592b023008c8bb261bc5bba0c6e2
https://hal.inria.fr/hal-00772046

Zobrazit plný text záznamu

Vyhledávací nástroje:

Upřesnit hledání