Michael Gimelfarb

reinforcement learning (3) transfer learning (2) bayesian model combination (2) value function (1) value iteration (1) policy improvement (1) symbolic dynamic programming (1) mixture of expert (1) utility function (1) successor feature (1) model-free reinforcement learning (1) reward shaping (1) q-value function (1) policy transfer (1) contextual reinforcement learning (1) bayesian mixture model (1) exploration-exploitation trade-off (1) multi-expert learning (1) model-free learning (1) deep q-learning (1)