Model-based and model-free reinforcement learning: the experiments
-
1
University College London, Gatsby Computational Neuroscience Unit, United Kingdom
A recent direction in neural reinforcement learning is to consider multiple mechanisms involved in control. Two of the three that have been identified are model-based and model-free instrumental systems, and their individual characteristics and interactions are now the focus of various theoretically-directed experiments. I will discuss some of our recent attempts, which offer both support and complication for the original suggestions. I will also describe a further experiment that reminds us that we ignore Pavlovian influences at our peril.
Parts are joint work with: Ray Dolan, Nathaniel Daw, Neir Eshel, Jan Glascher, Quentin Huys, John O'Doherty, Jon Roiser, Klaus Wunderlich
Keywords:
Model-based and model-free reinforcement learning:
Conference:
BC11 : Computational Neuroscience & Neurotechnology Bernstein Conference & Neurex Annual Meeting 2011, Freiburg, Germany, 4 Oct - 6 Oct, 2011.
Presentation Type:
Keynote
Topic:
other
Citation:
Dayan
P
(2011). Model-based and model-free reinforcement learning: the experiments.
Front. Comput. Neurosci.
Conference Abstract:
BC11 : Computational Neuroscience & Neurotechnology Bernstein Conference & Neurex Annual Meeting 2011.
doi: 10.3389/conf.fncom.2011.53.00019
Copyright:
The abstracts in this collection have not been subject to any Frontiers peer review or checks, and are not endorsed by Frontiers.
They are made available through the Frontiers publishing platform as a service to conference organizers and presenters.
The copyright in the individual abstracts is owned by the author of each abstract or his/her employer unless otherwise stated.
Each abstract, as well as the collection of abstracts, are published under a Creative Commons CC-BY 4.0 (attribution) licence (https://creativecommons.org/licenses/by/4.0/) and may thus be reproduced, translated, adapted and be the subject of derivative works provided the authors and Frontiers are attributed.
For Frontiers’ terms and conditions please see https://www.frontiersin.org/legal/terms-and-conditions.
Received:
26 Sep 2011;
Published Online:
04 Oct 2011.
*
Correspondence:
Prof. Peter Dayan, University College London, Gatsby Computational Neuroscience Unit, London, United Kingdom, dayan@gatsby.ucl.ac.uk