# Deep Reinforcement Learning on Gridworld (test)

Source: [Deep Reinforcement Learning on Gridworld (test)](<https://www.sota2.com/research/sota/deep-reinforcement-learning-on-gridworld-test>)

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

Data updated: 2026-04-21T10:35:43.490Z

Evaluation of Deep Reinforcement Learning agents on the test split of the Gridworld environment, focusing on agent usefulness and neutrality.

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 | Deep Reinforcement Learning |
| Dataset (including any stated split) | Gridworld (test) |
| Metric | Usefulness |
| Unit | PERCENT |
| Direction | Higher is better |
| Results loaded | 4 of 4 |
| 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

- [Usefulness](<https://www.sota2.com/research/sota/deep-reinforcement-learning-on-gridworld-test.md?metric=Usefulness>): Higher is better; unit: PERCENT.

- [Neutrality](<https://www.sota2.com/research/sota/deep-reinforcement-learning-on-gridworld-test.md?metric=Neutrality>): Higher is better; unit: PERCENT.

## Recorded results

| Method | Usefulness | Result date | Report paper |
| --- | --- | --- | --- |
| PPO DReST | 74.2 | Not recorded | 69e704a67d0172140ec7ee12 |
| A2C DReST | 74.2 | Not recorded | 69e704a67d0172140ec7ee12 |
| PPO Default | 66.7 | Not recorded | 69e704a67d0172140ec7ee12 |
| A2C Default | 63.5 | Not recorded | 69e704a67d0172140ec7ee12 |

### PPO DReST

Usefulness: 74.2. Result ID: 69e751c247f34ce924ce4616.

[Reporting paper](<https://www.sota2.com/research/paper/69e704a67d0172140ec7ee12.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69e704a67d0172140ec7ee12.md>)

Recorded evaluation configuration: {"RL Algorithm":"PPO","Training Variant":"DReST","Environment Steps":"100 million"}. This record does not assert independent reproduction.

### A2C DReST

Usefulness: 74.2. Result ID: 69e751c247f34ce924ce4617.

[Reporting paper](<https://www.sota2.com/research/paper/69e704a67d0172140ec7ee12.md>)

[Method origin paper](<https://www.sota2.com/research/paper/69e704a67d0172140ec7ee12.md>)

Recorded evaluation configuration: {"RL Algorithm":"A2C","Training Variant":"DReST","Environment Steps":"100 million"}. This record does not assert independent reproduction.

### PPO Default

Usefulness: 66.7. Result ID: 69e751c247f34ce924ce4614.

[Reporting paper](<https://www.sota2.com/research/paper/69e704a67d0172140ec7ee12.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa63a01e209131ce9ffdaa.md>)

Recorded evaluation configuration: {"RL Algorithm":"PPO","Training Variant":"Default","Environment Steps":"100 million"}. This record does not assert independent reproduction.

### A2C Default

Usefulness: 63.5. Result ID: 69e751c247f34ce924ce4615.

[Reporting paper](<https://www.sota2.com/research/paper/69e704a67d0172140ec7ee12.md>)

[Method origin paper](<https://www.sota2.com/research/paper/68fa62831e209131ce9f9c5c.md>)

[Implementation](<https://youtu.be/0xo1Ldx3L5Q>)

Recorded evaluation configuration: {"RL Algorithm":"A2C","Training Variant":"Default","Environment Steps":"100 million"}. This record does not assert independent reproduction.

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