Zhong, V;
Mu, J;
Zettlemoyer, L;
Grefenstette, E;
Rocktäschel, T;
(2022)
Improving Policy Learning via Language Dynamics Distillation.
In:
Advances in Neural Information Processing Systems.
NeurIPS
Preview |
Text
3780_improving_policy_learning_via_.pdf - Published Version Download (6MB) | Preview |
Abstract
Recent work has shown that augmenting environments with language descriptions improves policy learning. However, for environments with complex language abstractions, learning how to ground language to observations is difficult due to sparse, delayed rewards. We propose Language Dynamics Distillation (LDD), which pretrains a model to predict environment dynamics given demonstrations with language descriptions, and then fine-tunes these language-aware pretrained representations via reinforcement learning (RL). In this way, the model is trained to both maximize expected reward and retain knowledge about how language relates to environment dynamics. On SILG, a benchmark of five tasks with language descriptions that evaluate distinct generalization challenges on unseen environments (NetHack, ALFWorld, RTFM, Messenger, and Touchdown), LDD outperforms tabula-rasa RL, VAE pretraining, and methods that learn from unlabeled demonstrations in inverse RL and reward shaping with pretrained experts. In our analyses, we show that language descriptions in demonstrations improve sample-efficiency and generalization across environments, and that dynamics modeling with expert demonstrations is more effective than with non-experts.
| Type: | Proceedings paper |
|---|---|
| Title: | Improving Policy Learning via Language Dynamics Distillation |
| Event: | 36th Conference on Neural Information Processing Systems (NeurIPS 2022) |
| ISBN-13: | 9781713871088 |
| Open access status: | An open access version is available from UCL Discovery |
| Publisher version: | https://proceedings.neurips.cc/paper_files/paper/2... |
| 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. |
| UCL classification: | UCL UCL > Provost and Vice Provost Offices > UCL BEAMS UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science UCL > Provost and Vice Provost Offices > UCL BEAMS > Faculty of Engineering Science > Dept of Computer Science |
| URI: | https://discovery.ucl.ac.uk/id/eprint/10173695 |
Loading...
Loading...Loading...| 1. |  China | 1 |
Archive Staff Only
 |
View Item |
