作者: N. Mastronarde , M. van der Schaar
关键词: Artificial intelligence 、 Multimedia 、 Stability (learning theory) 、 Robot learning 、 Instance-based learning 、 Probabilistic logic 、 Competitive learning 、 Markov process 、 Computational learning theory 、 Markov decision process 、 Online machine learning 、 Active learning (machine learning) 、 Reinforcement learning 、 Computer science
摘要: … In Section VI-D, we show that the learning performance (including the weighted estimation error and the average reward) can be dramatically improved by smartly updating multiple state…