Rui Kong

policy optimization (3) offline reinforcement learning (2) action selection (2) reinforcement learning (1) value function (1) policy gradient (1) question answering (1) data augmentation (1) lower bound (1) dynamic routing (1) partially observable markov decision process (1) dynamic programming (1) online planning (1) value iteration (1) preference alignment (1) continual learning (1) policy improvement (1) kernel optimization (1) sequence modeling (1) model compression (1)