如何学习良好的潜在表示是现代机器学习时代的一个重要课题。对于强化学习,使用一个好的表示使决策过程更加有效。本次演讲,我将介绍我们的工作,构建基于任务的潜在操作空间,用于基于搜索的黑盒函数优化,寻找策略变更的表示,该表示支持在不完全信息协同博弈中联合策略搜索,以及不同的表示如何影响RL探索。
视频:
https://www.youtube.com/watch?v=sH4a2a0ntUA
Yuandong Tian is a Research Scientist and Manager in Facebook AI Research, working on deep reinforcement learning, representation learning and optimization. He is the lead scientist and engineer for ELF OpenGo and DarkForest Go projects. Prior to that, he was in Google Self-driving Car team in 2013-2014. He received a Ph.D in Robotics Institute, Carnegie Mellon University in 2013. He is the recipient of 2013 ICCV Marr Prize Honorable Mentions.
专知便捷查看
便捷下载,请关注专知公众号(点击上方蓝色专知关注)
后台回复“RL71” 就可以获取《【Facebook-Yuandong Tian】在RL中为搜索和探索找到良好的表示,附71页PPT与视频》专知下载链接