Moskovitz, T;
Kao, C;
Sahani, M;
Botvinick, MM;
(2023)
Minimum Description Length Control.
In:
11th International Conference on Learning Representations, ICLR 2023.
(pp. pp. 1-31).
ICLR (International Conference on Learning Representations)
Preview |
Text
3218_minimum_description_length_con.pdf - Published Version Download (9MB) | Preview |
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.
| Type: | Proceedings paper |
|---|---|
| Title: | Minimum Description Length Control |
| Open access status: | An open access version is available from UCL Discovery |
| Publisher version: | https://openreview.net/forum?id=oX3tGygjW1q |
| Language: | English |
| Additional information: | This version is the version of record. For information on re-use, please refer to the publisher's terms and conditions. |
| Keywords: | multitask reinforcement learning, RL, reinforcement learning, MDL |
| UCL classification: | UCL UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Life Sciences UCL > Provost and Vice Provost Offices > School of Life and Medical Sciences > Faculty of Life Sciences > Gatsby Computational Neurosci Unit |
| URI: | https://discovery.ucl.ac.uk/id/eprint/10195574 |
Archive Staff Only
 |
View Item |
