Invariant causal prediction for block MDPs
Generalization across environments is critical to the successful application of reinforcement learning (RL) algorithms to real-world challenges. In this work we propose a method for learning state abstractions which generalize to novel observation distributions in the multi-environment RL setting. W...
| প্রধান লেখক: | , , , , , , , |
|---|---|
| বিন্যাস: | Conference item |
| ভাষা: | English |
| প্রকাশিত: |
Proceedings of Machine Learning Research
2020
|