What it does
The reported single-seed pilot found that Jev's frozen risk features did not help PPO reach the target sooner—a useful negative result rather than a claimed win.
Benchmarks & research
mario-play
An original platformer and reinforcement-learning lab testing whether frozen Jev risk features help PPO learn faster.
Bookmark: mario-play
Keep this in your collection.
Leave a noteWhat would you try with this?
: mario-play
Only you can see your notes.
Screenshot unavailable. Open the experiment ↗
How you can use it
A developer can adapt this project to let an AI play a video game. The software sends what happens on the screen to TypeSafe. This outside service then picks the next move, like running or jumping.
To run the AI live, the developer must provide a TypeSafe access key. This key connects the game to the outside service. The project also includes automated players that do not need a key. These test scripts follow fixed rules to play on their own.