DifferentiableEvolutionaryRL/DERL-ALFWorld-L2-Qwen2.5-1.5B
2B
•
Updated
•
28
None defined yet.
This repo presents the models of the paper Differentiable Evolutionary Reinforcement Learning. The models are trained by our bi-level evolutionary training loop.
We release the base initial Meta-Optimizer and the best policy model for each task.