In-Context Reinforcement Learning for Variable Action Spaces

December 20, 2023 ยท Entered Twilight ยท ๐Ÿ› International Conference on Machine Learning

๐Ÿ’ค TWILIGHT: Eternal Rest
Repo abandoned since publication

Repo contents: .gitignore, Dockerfile, README.md, configs, envs, reports, requirements.txt, src, sweeps, tiny_llama_requirements.txt

Authors Viacheslav Sinii, Alexander Nikulin, Vladislav Kurenkov, Ilya Zisman, Sergey Kolesnikov arXiv ID 2312.13327 Category cs.LG: Machine Learning Cross-listed cs.AI Citations 25 Venue International Conference on Machine Learning Repository https://github.com/corl-team/headless-ad โญ 91 Last Checked 1 month ago
Abstract
Recently, it has been shown that transformers pre-trained on diverse datasets with multi-episode contexts can generalize to new reinforcement learning tasks in-context. A key limitation of previously proposed models is their reliance on a predefined action space size and structure. The introduction of a new action space often requires data re-collection and model re-training, which can be costly for some applications. In our work, we show that it is possible to mitigate this issue by proposing the Headless-AD model that, despite being trained only once, is capable of generalizing to discrete action spaces of variable size, semantic content and order. By experimenting with Bernoulli and contextual bandits, as well as a gridworld environment, we show that Headless-AD exhibits significant capability to generalize to action spaces it has never encountered, even outperforming specialized models trained for a specific set of actions on several environment configurations. Implementation is available at: https://github.com/corl-team/headless-ad.
Community shame:
Not yet rated
Community Contributions

Found the code? Know the venue? Think something is wrong? Let us know!

๐Ÿ“œ Similar Papers

In the same crypt โ€” Machine Learning