Learning to Solve New sequential decision-making Tasks with In-Context Learning
Primary Area: transfer learning, meta learning, and lifelong learning
Code Of Ethics: I acknowledge that I and all co-authors of this work have read and commit to adhering to the ICLR Code of Ethics.
Keywords: In-Context Learning, Decision Making, Generalization
Submission Guidelines: I certify that this submission complies with the submission instructions as described on https://iclr.cc/Conferences/2024/AuthorGuide.
Abstract: Training autonomous agents that can generalize to new tasks from a small number of demonstrations is a long-standing problem in machine learning. Recently, transformers have displayed impressive few-shot learning capabilities on a wide range of domains in language and vision. However, the sequential decision-making setting poses additional challenges and has a much lower tolerance for errors since the environment's stochasticity or the agent's wrong actions can lead to unseen (and sometimes unrecoverable) states. In this paper, we use an illustrative example to show that a naive approach to using transformers in sequential decision-making problems does not lead to few-shot learning. We then demonstrate how training on sequences of trajectories with certain distributional properties leads to few-shot learning in new sequential decision-making tasks. We investigate different design choices and find that larger model and dataset sizes, as well as more task diversity, environment stochasticity and trajectory burstiness, all result in better in-context learning of new out-of-distribution tasks. Our work demonstrates that by leveraging large offline pretraining datasets, our model is able to generalize to unseen MiniHack and Procgen tasks via in-context learning, from just a handful of expert demonstrations per task.
Anonymous Url: I certify that there is no URL (e.g., github page) that could be used to find authors' identity.
No Acknowledgement Section: I certify that there is no acknowledgement section in this submission for double blind review.
Submission Number: 6225
Loading