Name
Abstract | ||
|---|---|---|
Deep reinforcement learning encompasses many versatile tools for designing learning agents that can perform well on a variety of high-dimensional visual tasks, ranging from video games to robotic manipulation. However, these methods typically suffer from poor sample efficiency, partially because they strive to be largely problem-agnostic. In this work, we demonstrate the utility of a different approach that is extremely sample efficient, but limited to object-centric tasks that (approximately) obey basic physical laws. Specifically, we propose the Hypothesis Proposal and Evaluation (HyPE) algorithm, which utilizes a small set of intuitive assumptions about the behavior of objects in the physical world (or in games that mimic physics) to automatically define and learn hierarchical skills in a highly efficient manner. HyPE does this by discovering objects from raw pixel data, generating hypotheses about the controllability of observed changes in object state, and learning a hierarchy of skills that can test these hypotheses and control increasingly complex interactions with objects. We demonstrate that HyPE can dramatically improve sample efficiency when learning a high-quality pixels-to-actions policy; in the popular benchmark task, Breakout, HyPE learns an order of magnitude faster than common baseline reinforcement learning and evolutionary strategies for policy learning. |
Year | DOI | Venue |
|---|---|---|
2020 | 10.1109/IROS45743.2020.9340891 | CoRR |
DocType | Volume | Citations |
Conference | abs/1906.01408 | 0 |
PageRank | References | Authors |
0.34 | 0 | 3 |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Caleb Chuck | 1 | 0 | 0.68 |
| Supawit Chockchowwat | 2 | 0 | 0.34 |
| S. Niekum | 3 | 165 | 23.73 |