[2009.05524] Physically Embedded Planning Problems: New Challenges for Reinforcement Learningopen searchopen navigation menucontact arXivsubscribe to arXiv mailings

Recent work in deep reinforcement learning (RL) has produced algorithms capable of mastering challenging games such as Go, chess, or shogi. In these works the RL agent directly observes the natural state of the game and controls that state directly with its actions. However, when humans play such games, they do not just reason about the moves but also interact with their physical environment. They understand the state of the game by looking at the physical board in front of them and modify it by manipulating pieces using touch and fine-grained motor control. Mastering complicated physical systems with abstract goals is a central challenge for artificial intelligence, but it remains out of reach for existing RL algorithms. To encourage progress towards this goal we introduce a set of physically embedded planning problems and make them publicly available. We embed challenging symbolic tasks (Sokoban, tic-tac-toe, and Go) in a physics engine to produce a set of tasks that require percepti

4 mentions: @hardmaru@animesh_garg@ak92501
Date: 2020/09/14 05:21

Referring Tweets

@hardmaru Someone is exploring this direction❗👌 Physically Embedded Planning Problems: New Challenges for RL “When humans play such games (such as Go, chess, or shogi), they do not just reason about the moves but also interact with their physical environment…” t.co/RmV1pIQVjE t.co/0oiz9yfoD5

Related Entries

Read more [2002.05638] GANILLA: Generative Adversarial Networks for Image to Illustration Translationcontact a...
0 users, 3 mentions 2020/02/16 12:54
Read more [1910.05253] Adversarial Colorization Of Icons Based On Structure And Color Conditionsopen searchope...
0 users, 3 mentions 2020/05/19 17:21
Read more [2007.02190] BézierSketch: A generative model for scalable vector sketchesopen searchopen navigation...
0 users, 2 mentions 2020/07/07 03:51
Read more Self-Supervised Policy Adaptation during Deployment
0 users, 5 mentions 2020/07/09 08:21
Read more [2009.08319] Decoupling Representation Learning from Reinforcement Learningopen searchopen navigatio...
0 users, 6 mentions 2020/09/18 08:21