Autor: |
Modhe, Nirbhay, Chattopadhyay, Prithvijit, Sharma, Mohit, Das, Abhishek, Parikh, Devi, Batra, Dhruv, Vedantam, Ramakrishna |
Rok vydání: |
2019 |
Předmět: |
|
Druh dokumentu: |
Working Paper |
DOI: |
10.24963/ijcai.2020/280 |
Popis: |
We propose a novel framework to identify sub-goals useful for exploration in sequential decision making tasks under partial observability. We utilize the variational intrinsic control framework (Gregor et.al., 2016) which maximizes empowerment -- the ability to reliably reach a diverse set of states and show how to identify sub-goals as states with high necessary option information through an information theoretic regularizer. Despite being discovered without explicit goal supervision, our sub-goals provide better exploration and sample complexity on challenging grid-world navigation tasks compared to supervised counterparts in prior work. |
Databáze: |
arXiv |
Externí odkaz: |
|