Vision-Language Models as a Source of Rewards
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , , , , , , , , , , , , , , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866909252775313408 |
|---|---|
| author | Baumli, Kate Baveja, Satinder Behbahani, Feryal Chan, Harris Comanici, Gheorghe Flennerhag, Sebastian Gazeau, Maxime Holsheimer, Kristian Horgan, Dan Laskin, Michael Lyle, Clare Masoom, Hussain McKinney, Kay Mnih, Volodymyr Neitz, Alexander Nikulin, Dmitry Pardo, Fabio Parker-Holder, Jack Quan, John Rocktäschel, Tim Sahni, Himanshu Schaul, Tom Schroecker, Yannick Spencer, Stephen Steigerwald, Richie Wang, Luyu Zhang, Lei |
| author_facet | Baumli, Kate Baveja, Satinder Behbahani, Feryal Chan, Harris Comanici, Gheorghe Flennerhag, Sebastian Gazeau, Maxime Holsheimer, Kristian Horgan, Dan Laskin, Michael Lyle, Clare Masoom, Hussain McKinney, Kay Mnih, Volodymyr Neitz, Alexander Nikulin, Dmitry Pardo, Fabio Parker-Holder, Jack Quan, John Rocktäschel, Tim Sahni, Himanshu Schaul, Tom Schroecker, Yannick Spencer, Stephen Steigerwald, Richie Wang, Luyu Zhang, Lei |
| contents | Building generalist agents that can accomplish many goals in rich open-ended environments is one of the research frontiers for reinforcement learning. A key limiting factor for building generalist agents with RL has been the need for a large number of reward functions for achieving different goals. We investigate the feasibility of using off-the-shelf vision-language models, or VLMs, as sources of rewards for reinforcement learning agents. We show how rewards for visual achievement of a variety of language goals can be derived from the CLIP family of models, and used to train RL agents that can achieve a variety of language goals. We showcase this approach in two distinct visual domains and present a scaling trend showing how larger VLMs lead to more accurate rewards for visual goal achievement, which in turn produces more capable RL agents. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2312_09187 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Vision-Language Models as a Source of Rewards Baumli, Kate Baveja, Satinder Behbahani, Feryal Chan, Harris Comanici, Gheorghe Flennerhag, Sebastian Gazeau, Maxime Holsheimer, Kristian Horgan, Dan Laskin, Michael Lyle, Clare Masoom, Hussain McKinney, Kay Mnih, Volodymyr Neitz, Alexander Nikulin, Dmitry Pardo, Fabio Parker-Holder, Jack Quan, John Rocktäschel, Tim Sahni, Himanshu Schaul, Tom Schroecker, Yannick Spencer, Stephen Steigerwald, Richie Wang, Luyu Zhang, Lei Machine Learning Building generalist agents that can accomplish many goals in rich open-ended environments is one of the research frontiers for reinforcement learning. A key limiting factor for building generalist agents with RL has been the need for a large number of reward functions for achieving different goals. We investigate the feasibility of using off-the-shelf vision-language models, or VLMs, as sources of rewards for reinforcement learning agents. We show how rewards for visual achievement of a variety of language goals can be derived from the CLIP family of models, and used to train RL agents that can achieve a variety of language goals. We showcase this approach in two distinct visual domains and present a scaling trend showing how larger VLMs lead to more accurate rewards for visual goal achievement, which in turn produces more capable RL agents. |
| title | Vision-Language Models as a Source of Rewards |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2312.09187 |