Multi-timestep models for Model-based Reinforcement Learning
This addresses a critical challenge in MBRL for improving long-term predictions, but it is incremental as it builds on existing one-step models with a modified objective.
The paper tackles the problem of compounding prediction errors in model-based reinforcement learning by training one-step models with a multi-timestep objective, resulting in improved long-horizon R2 scores and outperforming standard models in batch RL scenarios, especially with noisy data.
In model-based reinforcement learning (MBRL), most algorithms rely on simulating trajectories from one-step dynamics models learned on data. A critical challenge of this approach is the compounding of one-step prediction errors as length of the trajectory grows. In this paper we tackle this issue by using a multi-timestep objective to train one-step models. Our objective is a weighted sum of a loss function (e.g., negative log-likelihood) at various future horizons. We explore and test a range of weights profiles. We find that exponentially decaying weights lead to models that significantly improve the long-horizon R2 score. This improvement is particularly noticeable when the models were evaluated on noisy data. Finally, using a soft actor-critic (SAC) agent in pure batch reinforcement learning (RL) and iterated batch RL scenarios, we found that our multi-timestep models outperform or match standard one-step models. This was especially evident in a noisy variant of the considered environment, highlighting the potential of our approach in real-world applications.