---------- Forwarded message ---------- From: Doina Precup dprecup@cs.mcgill.ca Date: Fri, Oct 3, 2014 at 2:14 PM Subject: Damien Ernst talk, ***Wed October 8***, 2pm, MC 437 To: "labrl@cs.mcgill.ca" labrl@cs.mcgill.ca, Bengio Yoshua < bengioy@iro.umontreal.ca> Cc: "dernst@ulg.ac.be" dernst@ulg.ac.be
Hi everyone,
Damien Ernst (http://www.montefiore.ulg.ac.be/~ernst/) will be visiting us the coming Wed. He will give a talk at 2pm in MC 437, titled:
Batch Mode Reinforcement Learning based on the Synthesis of Artificial Trajectories
Abstract: We consider the batch mode reinforcement learning setting, where the central problem is to learn from a sample of trajectories a policy that satisfies or optimizes a performance criterion. We focus on the continuous state space case for which usual resolution schemes rely on function approximators either to represent the underlying control problem or to represent its value function. As an alternative to the use of function approximators, we rely on the synthesis of “artificial trajectories” from the given sample of trajectories, and show that this idea opens new avenues for designing and analyzing algorithms for batch mode reinforcement learning.
Best, Doina
Afficher les réponses par date