Operator Splitting Value Iteration
This work addresses planning and reinforcement learning efficiency for researchers and practitioners, representing an incremental improvement over existing methods like Dyna.
The paper tackles the problem of accelerating value function convergence in discounted MDPs by introducing Operator Splitting Value Iteration (OS-VI) and its sample-based version OS-Dyna, which achieve faster convergence with accurate models and maintain correctness despite model errors.
We introduce new planning and reinforcement learning algorithms for discounted MDPs that utilize an approximate model of the environment to accelerate the convergence of the value function. Inspired by the splitting approach in numerical linear algebra, we introduce Operator Splitting Value Iteration (OS-VI) for both Policy Evaluation and Control problems. OS-VI achieves a much faster convergence rate when the model is accurate enough. We also introduce a sample-based version of the algorithm called OS-Dyna. Unlike the traditional Dyna architecture, OS-Dyna still converges to the correct value function in presence of model approximation error.