# Epsilon-Optimal Policy Learning on Infinite-horizon Discounted CMDPs (Safe Exploration)

Source: [Epsilon-Optimal Policy Learning on Infinite-horizon Discounted CMDPs (Safe Exploration)](<https://www.sota2.com/research/sota/epsilon-optimal-policy-learning-on-infinite-horizon-discounted-cmdps-safe-exploration>)

Collection index: [llms.txt](<https://www.sota2.com/research/llms.txt>)

Data updated: 2026-02-26T05:21:57.210Z

Evaluation of algorithms designed to learn an epsilon-optimal policy while ensuring safe exploration guarantees within the theoretical framework of infinite-horizon discounted Constrained Markov Decision Processes (CMDPs), typically measured by sample complexity.

A research result applies to its stated dataset, split, metric and evaluation configuration. Compare like-for-like settings, preserve the metric's unit and direction, and distinguish paper-reported results from independent reproduction. The best result recorded here is not a claim of exhaustive, current global coverage.

| Scope | Recorded value |
| --- | --- |
| Task | Learning an epsilon-optimal policy with safe exploration |
| Dataset (including any stated split) | Infinite-horizon Discounted Constrained Markov Decision Processes (CMDPs) |
| Metric | Sample Complexity |
| Unit | SCALAR |
| Direction | Lower is better |
| Results loaded | 0 of 0 |
| Coverage | All results currently returned for this benchmark |

Evaluation settings may differ between results. The original papers and recorded configurations are needed to establish comparability. A shared metric name alone does not establish an identical protocol.

## Metrics

- [Sample Complexity](<https://www.sota2.com/research/sota/epsilon-optimal-policy-learning-on-infinite-horizon-discounted-cmdps-safe-exploration.md?metric=Sample%20Complexity>): Lower is better; unit: SCALAR.

## Recorded results

| Method | Sample Complexity | Result date | Report paper |
| --- | --- | --- | --- |

Page 1 of 1. This is an index page, not the full collection.
