search

HumanCompatibleAI / population-irl

(Experimental) Inverse reinforcement learning from trajectories generated by multiple agents with different (but correlated) rewards
MIT License
26 stars 2 forks source link

Visualization #7

Closed AdamGleave closed 6 years ago

AdamGleave commented 6 years ago

Would be useful to be able to visualize trajectories sampled from RL algorithms. We'll probably want to start by defining an RGB render() method for the grid world (currently it's just ANSI), then generate a video given a trajectory input.

Discordius commented 6 years ago

I think we can consider this done, though there are still some improvements to be made.

AdamGleave commented 6 years ago

Agreed