3 ms·
SKRL – Reinforcement Learning Library
- syntaxing 4y agoWhoa this looks super interesting. I like how it has non deep learning approaches. When I was playing around with different models, my Q-learning would often outperform DQN for simple environment like cartpole.