02/02/2021

Augmenting Policy Learning with Routines Discovered from a Single Demonstration

Zelin Zhao, Chuang Gan, Jiajun Wu, Xiaoxiao Guo, Joshua B. Tenenbaum

Keywords:

Abstract Paper Similar Papers

Abstract: Humans can abstract prior knowledge from very little data and use it to boost skill learning. In this paper, we propose routine-augmented policy learning (RAPL), which discovers routines composed of primitive actions from a single demonstration and uses discovered routines to augment policy learning. To discover routines from the demonstration, we first abstract routine candidates by identifying grammar over the demonstrated action trajectory. Then, the best routines measured by length and frequency are selected to form a routine library. We propose to learn policy simultaneously at primitive-level and routine-level with discovered routines, leveraging the temporal structure of routines. Our approach enables imitating expert behavior at multiple temporal scales for imitation learning and promotes reinforcement learning exploration. Extensive experiments on Atari games demonstrate that RAPL improves the state-of-the-art imitation learning method SQIL and reinforcement learning method A2C. Further, we show that discovered routines can generalize to unseen levels and difficulties on the CoinRun benchmark.

The video of this talk cannot be embedded. You can watch it here:
https://slideslive.com/38947875
(Link will open in new window)
 0
 0
 0
 0
  Share    
This is an embedded video. Talk and the respective paper are published at AAAI 2021 virtual conference. If you are one of the authors of the paper and want to manage your upload, see the question "My papertalk has been externally embedded..." in the FAQ section.

Comments

Post Comment
no comments yet
code of conduct: tbd Characters remaining: 140

Similar Papers

 0
 0
 0
 0
 3:21 
 1
 0
 0
 0
 5:52 
 0
 0
 0
 0
 15:45 
 0
 0
 0
 0
 6:49 
 0
 0
 0
 0
 15:04 
 0
 0
 0
 0
 15:11 
 0
 0
 0
 0
 10:34 
 0
 0
 0
 0
 5:02 
 0
 0
 0
 0
 5:01 
 0
 0
 0
 0
 5:12 
 0
 0
 0
 0
 15:19 
 0
 0
 0
 0
 12:41 
 0
 0
 0
 0
 15:06 
 0
 0
 0
 0
 2:47 
 0
 0
 0
 0
 14:47 
 0
 0
 0
 0
 4:55 
 0
 0
 0
 0
 4:53 
 0
 0
 0
 0
 14:17 
 0
 0
 0
 0
 3:09 
 0
 0
 0
 0
 3:23 
 0
 0
 0
 0
 10:39 
 0
 0
 0
 0
 14:46 
 0
 0
 0
 0
 14:21 
 0
 0
 0
 0
 10:26 
 0
 0
 0
 0
 13:59 
 0
 0
 0
 0
 13:04 
 0
 0
 0
 0
 5:04 
 0
 0
 0
 0
 13:50 
 0
 0
 0
 0
 5:30 
 0
 0
 0
 0
 5:06 
 1
 0
 0
 0
 5:21 
 0
 0
 0
 0
 5:07 
 0
 0
 0
 0
 12:54 
 0
 0
 0
 0
 5:35 
 0
 0
 0
 0
 10:01 
 0
 0
 0
 0
 3:19 
 0
 0
 0
 0
 14:27 
 0
 0
 0
 0
 7:25 
 0
 0
 0
 0
 5:47 
 0
 0
 0
 0
 9:11