Will SOTA for RL on Atari-57 include a large pretrained language/image/video model by 2024?
8
79
240
resolved Jan 3
Resolved
NO

It's fine if the training dataset for the large model contains the Atari environments, this isn't a question about sample efficiency.

I'll accept SOTA for either mean or median HNS.

The Atari-57 benchmark on papers with code: https://paperswithcode.com/sota/atari-games-on-atari-57

Get Ṁ1,000 play money

🏅 Top traders

#NameTotal profit
1Ṁ32
2Ṁ5
3Ṁ1
Sort by:

Current SOTA is LBC, which does not use such a model. Resolving NO

What are you counting as SOTA? Top score on one game? Multiple games? All of them? A tie for top score on a game where it's fairly easy to get a perfect score, like Pong?

@Multicore Atari-57. I'll update the question

Will SOTA for RL on Atari environments include a large pretrained language/image/video model by 2024?, 8k, beautiful, illustration, trending on art station, picture of the day, epic composition