Markov flow policy -- deep MC
Fuente:
arXiv
Saved in:
| Main Authors: | , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866912007923433472 |
|---|---|
| author | Soffair, Nitsan Katz, Gilad |
| author_facet | Soffair, Nitsan Katz, Gilad |
| contents | Discounted algorithms often encounter evaluation errors due to their reliance on short-term estimations, which can impede their efficacy in addressing simple, short-term tasks and impose undesired temporal discounts (\(γ\)). Interestingly, these algorithms are often tested without applying a discount, a phenomenon we refer as the \textit{train-test bias}. In response to these challenges, we propose the Markov Flow Policy, which utilizes a non-negative neural network flow to enable comprehensive forward-view predictions. Through integration into the TD7 codebase and evaluation using the MuJoCo benchmark, we observe significant performance improvements, positioning MFP as a straightforward, practical, and easily implementable solution within the domain of average rewards algorithms. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2405_00877 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Markov flow policy -- deep MC Soffair, Nitsan Katz, Gilad Machine Learning Artificial Intelligence Discounted algorithms often encounter evaluation errors due to their reliance on short-term estimations, which can impede their efficacy in addressing simple, short-term tasks and impose undesired temporal discounts (\(γ\)). Interestingly, these algorithms are often tested without applying a discount, a phenomenon we refer as the \textit{train-test bias}. In response to these challenges, we propose the Markov Flow Policy, which utilizes a non-negative neural network flow to enable comprehensive forward-view predictions. Through integration into the TD7 codebase and evaluation using the MuJoCo benchmark, we observe significant performance improvements, positioning MFP as a straightforward, practical, and easily implementable solution within the domain of average rewards algorithms. |
| title | Markov flow policy -- deep MC |
| topic | Machine Learning Artificial Intelligence |
| url | https://arxiv.org/abs/2405.00877 |