Minimum Description Length Control

Abstract

We propose a novel framework for multitask reinforcement learning based on the minimum description length (MDL) principle. In this approach, which we term MDL-control (MDL-C), the agent learns the common structure among the tasks with which it is faced and then distills it into a simpler representation which facilitates faster convergence and generalization to new tasks. In doing so, MDL-C naturally balances adaptation to each task with epistemic uncertainty about the task distribution. We motivate MDL-C via formal connections between the MDL principle and Bayesian inference, derive theoretical performance guarantees, and demonstrate MDL-C's empirical effectiveness on both discrete and high-dimensional continuous control tasks.

Cite

Text

Moskovitz et al. "Minimum Description Length Control." International Conference on Learning Representations, 2023.

Markdown

[Moskovitz et al. "Minimum Description Length Control." International Conference on Learning Representations, 2023.](https://mlanthology.org/iclr/2023/moskovitz2023iclr-minimum/)

BibTeX

@inproceedings{moskovitz2023iclr-minimum,
  title     = {{Minimum Description Length Control}},
  author    = {Moskovitz, Ted and Kao, Ta-Chu and Sahani, Maneesh and Botvinick, Matthew},
  booktitle = {International Conference on Learning Representations},
  year      = {2023},
  url       = {https://mlanthology.org/iclr/2023/moskovitz2023iclr-minimum/}
}