[CTN] CTN seminar: Marc Bellemare, University of Alberta, Apr 23, 3.30pm in PAS 2464

[Matthijs van der Meer]

Dear all,

Please join us for next Tuesday's CTN seminar (Apr 23) by Marc
Bellemare, from the University of Alberta. Title and abstract follow below.

Time and place are the usual, 3.30pm on Tuesday in PAS 2464.

If you would like to meet with Marc, and/or come to dinner
(on Tuesday evening), please let me know.

Hope to see you there,

- Matt


Marc Bellemare
PhD Candidate, Computing Science
University of Alberta

http://webdocs.cs.ualberta.ca/~mg17/


Title: Pixels and Priors: Learning a Generative Model of Atari 2600 Games

The model-based approach to reinforcement is appealing as it allows the
agent to reason about unexplored parts of its environment. However, most
existing model learning techniques impose harsh restrictions on the
model structure or simply do not scale well to large domains. In this
presentation, I will present a new algorithm for learning generative
models of arbitrary Atari 2600 games. Using statistical data compression
methods, this algorithm achieves a per-step cost linear in the number of
pixels, making it an interesting alternative to other model learning
methods. One of the novel components of this algorithm is the quad-tree
factorization, which uses Bayesian model averaging to efficiently learn
a variable-resolution factorization of the Atari 2600 observation space.
I will provide redundancy bounds that guarantee the statistical
efficiency and soundness of the approach and some recent empirical
results on a variety of Atari 2600 games.