Double Q(σ) and Q(σ, λ): Unifying Reinforcement Learning Control Algorithms

This repository includes code for the paper:

Double Q(σ) and Q(σ, λ): Unifying Reinforcement Learning Control Algorithms

It contains python implementations of Double Q(σ) and Q(σ, λ). These algorithms generalize Q-Learning, Sarsa and Q(σ) and can achieve better performance. Read the paper at arXiv for more information.