Reward-Predictive Clustering
Lehnert, Lucas, Frank, Michael J., Littman, Michael L.
–arXiv.org Artificial Intelligence
Recent advances in reinforcement-learning research have demonstrated impressive results in building algorithms that can out-perform humans in complex tasks. Nevertheless, creating reinforcement-learning systems that can build abstractions of their experience to accelerate learning in new contexts still remains an active area of research. Previous work showed that reward-predictive state abstractions fulfill this goal, but have only be applied to tabular settings. Here, we provide a clustering algorithm that enables the application of such state abstractions to deep learning settings, providing compressed representations of an agent's inputs that preserve the ability to predict sequences of reward. A convergence theorem and simulations show that the resulting reward-predictive deep network maximally compresses the agent's inputs, significantly speeding up learning in high dimensional visual control tasks. Furthermore, we present different generalization experiments and analyze under which conditions a pre-trained reward-predictive representation network can be re-used without re-training to accelerate learning -- a form of systematic out-of-distribution transfer.
arXiv.org Artificial Intelligence
Nov-6-2022
- Country:
- North America
- United States > New York
- New York County > New York City (0.04)
- Canada > Quebec
- Montreal (0.04)
- United States > New York
- Europe
- North America
- Genre:
- Research Report > New Finding (0.34)
- Industry:
- Health & Medicine (0.67)
- Leisure & Entertainment > Games (0.46)
- Education (0.46)
- Technology: