torch_rl

Reinforcement learning algorithms for spiking networks and artificial neural networks.

Currently implemented

1. Deep deterministic policy gradients with hindsight experience replay
2. Stochastic policy gradient with hindsight experience replay
3. Biased hindsight policy gradient
4. Proximal Policy Optimization on GPU
5. Covariance Matrix Adaptation Evolutionary Strategy

In progress...

1. Distributed proximal policy optimization