Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

More comprehensive Atari data #6

Open
void-elf opened this issue May 23, 2017 · 4 comments
Open

More comprehensive Atari data #6

void-elf opened this issue May 23, 2017 · 4 comments

Comments

@void-elf
Copy link
Contributor

void-elf commented May 23, 2017

We have quite a bit of Atari data, but for many of the games we only have 1-3 data points, all from one paper.

There is a lot to be added from the papers listed in https://github.com/AI-metrics/master_text/blob/master/AI-metrics-data.txt and by looking for more recent results (especially for Montezuma's Revenge)

This would be a great task for a volunteer just getting started with this project.

@pde
Copy link
Member

pde commented May 31, 2017

This looks almost done! The only additional thing I know we should do is make sure we've included all of these papers and then we can mark this as done.

@pde pde assigned void-elf and unassigned void-elf May 31, 2017
@pde pde added the data label Jun 21, 2017
@pde
Copy link
Member

pde commented Jun 21, 2017

The OpenAI Evolutionary Strategies paper has results on lots of Atari games, and got the state of the art on quite a few of them. So we should figure out how to transfer thei entire results table.

@pde
Copy link
Member

pde commented Jun 21, 2017

(Also that paper is interesting since it retrains A3C; we should use algorithm_src="https://arxiv.org/abs/1602.01783" if we add those results)

@pde
Copy link
Member

pde commented Jun 28, 2017

The Evolutionary Strategies data seemed best handled with a simple scraper, so I wrote one

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Projects
None yet
Development

No branches or pull requests

2 participants