arXiv
Open Access
2023
When do discounted-optimal policies also optimize the gain?
Victor Boone
Abstrak
In this technical note, we establish an upper-bound on the threshold on the discount factor starting from which all discounted-optimal deterministic policies are gain-optimal, that we prove to be tight on an example. To address computability issues of that theoretical threshold, we provide a weaker bound which is tractable on ergodic MDPs in polynomial time.
Penulis (1)
V
Victor Boone
Akses Cepat
Informasi Jurnal
- Tahun Terbit
- 2023
- Bahasa
- en
- Sumber Database
- arXiv
- Akses
- Open Access ✓