Name
Abstract | ||
|---|---|---|
Imitation learning allows agents to learn complex behaviors from demonstrations. However, learning a complex vision-based task may require an impractical number of demonstrations. Meta-imitation learning is a promising approach towards enabling agents to learn a new task from one or a few demonstrations by leveraging experience from learning similar tasks. In the presence of task ambiguity or unobserved dynamics, demonstrations alone may not provide enough information; an agent must also try the task to successfully infer a policy. In this work, we propose a method that can learn to learn from both demonstrations and trial-and-error experience with sparse reward feedback. In comparison to meta-imitation, this approach enables the agent to effectively and efficiently improve itself autonomously beyond the demonstration data. In comparison to meta-reinforcement learning, we can scale to substantially broader distributions of tasks, as the demonstration reduces the burden of exploration. Our experiments show that our method significantly outperforms prior approaches on a set of challenging, vision-based control tasks. |
Year | Venue | Keywords |
|---|---|---|
2020 | ICLR | meta-learning, reinforcement learning, imitation learning |
DocType | Citations | PageRank |
Conference | 0 | 0.34 |
References | Authors | |
25 | 10 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| Allan Zhou | 1 | 0 | 0.34 |
| Eric Jang | 2 | 88 | 4.90 |
| Daniel Kappler | 3 | 0 | 1.01 |
| Alex Herzog | 4 | 0 | 0.34 |
| Seyed Mohammad Khansari-Zadeh | 5 | 160 | 11.57 |
| Paul Wohlhart | 6 | 0 | 0.34 |
| Yunfei Bai | 7 | 92 | 9.48 |
| Mrinal Kalakrishnan | 8 | 636 | 33.36 |
| Sergey Levine | 9 | 3377 | 182.21 |
| Chelsea Finn | 10 | 819 | 57.17 |