Home » Authors » Mohammad Gheshlaghi Azar, Rémi Munos, Mohammad Ghavamzadeh, Hilbert Kappen

Mohammad Gheshlaghi Azar, Rémi Munos, Mohammad Ghavamzadeh, Hilbert Kappen

Public reference bibliography and cataloged PDF materials authored or published by Mohammad Gheshlaghi Azar, Rémi Munos, Mohammad Ghavamzadeh, Hilbert Kappen.

Found 1 document(s) Page 1 of 1

Reinforcement Learning with a Near Optimal Rate of Convergence

Year: 2011 • Source: hal.science • Listed PDF

We consider the problem of model-free reinforcement learning in the Markovian decision processes (MDP) under the PAC ("probably approximately correct") model. We introduce a new variant of Q-learning, called speedy Q-lea...