Michal Valko : Research

Michal Valko, machine learning scientist at DeepMind, Inria, and a lecturer at MVA/ENS PS.

  deep reinforcement learning, Monte-Carlo tree search, representation learning, active learning, graphs, bandit theory

News

News: new Apply for DeepMind Paris interenships for Summer 2022.
News: new Five papers accepted to NeurIPS 2021, including 1 oral and 2 spotlights!
News: new Congrats to Xuedong Shang for defending his thesis on Sep 29th, 2021!
News: new BraVe, self-supervised learning framework for video, accepted to ICCV 2021!
News: new IXOMD invited to be presented as RL Theory seminar!
News: new Our minimax SSP work invited to be presented as RL Theory seminar!
News: new Six papers accepted to ICML 2021, including the long talk on UCBMQ!
News: new KeRNS algorithm for non-stationary kernel RL accepted to AISTATS 2021!
News: new Three papers on reinforcement learning theory accepted to ALT 2021!
News: new After 4 years, the full Spectral Bandits paper with the lower bound and the comprehensive set of experiment is now online.
News: new I will be giving 3 online talks on BYOL in Novemebr and December 2020..
News: new Congrats to Julien Seznec for defending his thesis on Dec 15th, 2020!
News: new Congrats to Pierre Perrault for defending his thesis on Nov 30th, 2020!
News: new The Graphs in ML MVA course will start on January 5th, 2021 and will be taught by Daniele Calandriello.
News: new Five papers accepted to NeurIPS 2020 including two oral talks for BYOL and DISCO and 1 spotlight!.
News: new I am serving as an area chair for ICLR 2021.
News: very hot news Yannic Kilcher made a youtube video about our BYOL work!
News: very hot news Three months of lockdown lead to our three months intense self-supervised learning :-). BYOL is out!.
News: Eight papers accepted to ICML 2020. "See" you in Vienna!

older news

Bio

Michal is a machine learning scientist in DeepMind Paris, SequeL team at Inria, and the lecturer of the master course Graphs in Machine Learning at l'ENS Paris-Saclay. Michal is primarily interested in designing algorithms that would require as little human supervision as possible. This means 1) reducing the “intelligence” that humans need to input into the system and 2) minimizing the data that humans need to spend inspecting, classifying, or “tuning” the algorithms. Another important feature of machine learning algorithms should be the ability to adapt to changing environments. That is why he is working in domains that are able to deal with minimal feedback, such as online learning, bandit algorithms, semi-supervised learning, and anomaly detection. Most recently he has worked on sequential algorithms with structured decisions where exploiting the structure leads to provably faster learning. Structured learning requires more time and space resources and therefore the most recent work of Michal includes efficient approximations such as graph and matrix sketching with learning guarantees. In past, the common thread of Michal's work has been adaptive graph-based learning and its application to real-world applications such as recommender systems, medical error detection, and face recognition. His industrial collaborators include Adobe, Intel, Technicolor, and Microsoft Research. He received his Ph.D. in 2011 from the University of Pittsburgh under the supervision of Miloš Hauskrecht and after was a postdoc of Rémi Munos before taking a permanent position at Inria in 2012.

Collaborative Projects

  • CompLACS (EU FP7) - COMposing Learning for Artificial Cognitive Systems, 2011 - 2015 (with J. Shawe-Taylor)
  • DELTA (EU CHIST-ERA) - PC - Dynamically Evolving Long-Term Autonomy, 2018 - 2021 (with A. Jonsson)
  • PGMO-IRMO grant of Fondation Mathématique Jacques Hadamard: Theoretically grounded efficient algorithms for high-dimensional and continuous reinforcement learning, 2018 - 2020 (with M. Pirotta)
  • BoB (ANR) - Bayesian statistics for expensive models and tall data, 2016 - 2020 (with R. Bardenet)
  • LeLivreScolaire.fr - Sequential Learning for Educational Systems, 2017-2020 (PI)
  • BOLD (ANR) - PI - Beyond Online Learning for better Decision making, 2019 - 2023 (with V. Perchet)
  • Allocate - PI - Adaptive allocation of resources for recommender systems with U. Potsdam, 2017 - 2019 (with A. Carpentier)
  • INTEL/Inria - PI - Algorithmic Determination of IoT Edge Analytics Requirements, 2013 - 2014
  • Extra-Learn (ANR) - PI - EXtraction and TRAnsfer of knowledge in reinforcement LEARNing, 2014 - 2018 (with A. Lazaric) Lampada (ANR) - Learning Algorithms, Models an sPArse representations for structured DAta
  • NIH/NIGMS - R01 - Detecting deviations in clinical care in ICU data streams
  • NIH/NLM - R01 - Using medical records repositories to improve the alert system design
  • Inria/CWI – Sequential prediction & Understanding Deep RL, postdoc funding (PC, 2016-2018)
  • EduBand - coPI - Educational Bandits project with Carnegie Mellon, 2015 - 2018 (with A. Lazaric and E. Brunskill)

Students and postdocs

  • David Cheikhi, 2020 - 2021, Columbia Universitu, NYC/École Polytechnique, Paris, with Pierre Ménard
  • Robert Müller, 2020, Technical University of Munich, M2 student, with Pierre Ménard
  • Ahmed Choukarah, 2020, ENS Ulm, L3 student, with Pierre Ménard
  • Côme Fiegel, 2019, ENS Ulm, L3 student, with Victor Gabillon
  • Axel Elaldi, 2017-2018, master student, École Centrale de Lille ↝ ENS Paris-Saclay/MVA
  • Xuedong Shang, 2017, master student, ENS Rennes, with Emilie Kaufmann ↝ Inria
  • Guillaume Gautier, 2016, master student, École Normale Supérieure, Paris-Saclay, with Rémi Bardenet ↝ Inria/CNRS
  • Andrea Locatelli, 2015-2016, ENSAM/ENS Paris-Saclay, with Alexandra Carpentier ↝ Universität Potsdam
  • Souhail Toumdi, 2015 - 2016, master student, École Centrale de Lille, with Rémi Bardenet ↝ ENS Paris-Saclay/MVA
  • Akram Erraqabi, 2015, master student, École Polytechnique, Paris ↝ Université de Montréal
  • Mastane Achab, 2015, master student, École Polytechnique, Paris, with G. Neu ↝ l'ENS Paris-Saclay ↝ Télécom ParisTech
  • Jean-Bastien Grill, 2014, master student, École Normale Supérieure, Paris, with Rémi Munos ↝ Inria
  • Alexandre Dubus, 2012-2013, master student, Université Lille1 - Sciences et Technologies ↝ Inria
  • Karim Jedda, 2012-2013, master student, École Centrale de Lille ↝ ProSiebenSat.1
  • Alexis Wehrli, 2012-2013, master student, École Centrale de Lille ↝ ERDF

Contact

  • DeepMind Paris (bureau: FR-PAR-14L-2-205D)
  • 14 Rue de Londres
  • 75009 Paris
  • Inria Lille - Nord Europe, equipe SequeL (bureau: A05)
  • Parc Scientifique de la Haute Borne
  • 40 avenue Halley
  • 59650 Villeneuve d'Ascq, France
  • office phone: +33 3 59 57 7801
  • CMLA, ENS Paris-Saclay (bureau: vacataires)
  • 61 avenue du président Wilson
  • 40 avenue Halley
  • 94235 Cachan cedex


mv