levdoom

Here is 1 public repository matching this topic...

william25885 / levdoom-scene-specialized-dqn

Multi-level generalization on the LevDoom Seek and Slay benchmark: five per-level DQN experts routed by a learned scene classifier, compressed to fit a 51 MB deployment budget.

reinforcement-learning deep-reinforcement-learning rainbow pytorch dqn vizdoom generalization mixture-of-experts curiosity-driven-exploration levdoom

Updated Aug 19, 2026
Python

Add this topic to your repo

To associate your repository with the levdoom topic, visit your repo's landing page and select "manage topics."

Learn more