EVaDE : Event-Based Variational Thompson Sampling for Model-Based Reinforcement Learning
This work addresses the problem of efficient exploration in model-based reinforcement learning for researchers and practitioners, presenting an incremental improvement by adapting variational methods to object-based domains.
The paper tackles the challenge of approximating posterior distributions for model-based reinforcement learning in high-dimensional spaces by proposing Event-based Variational Distributions for Exploration (EVaDE), which uses event-based convolutional layers with Gaussian dropouts to direct exploration in object-based domains, achieving effectiveness on the 100K Atari game suite.
Posterior Sampling for Reinforcement Learning (PSRL) is a well-known algorithm that augments model-based reinforcement learning (MBRL) algorithms with Thompson sampling. PSRL maintains posterior distributions of the environment transition dynamics and the reward function, which are intractable for tasks with high-dimensional state and action spaces. Recent works show that dropout, used in conjunction with neural networks, induces variational distributions that can approximate these posteriors. In this paper, we propose Event-based Variational Distributions for Exploration (EVaDE), which are variational distributions that are useful for MBRL, especially when the underlying domain is object-based. We leverage the general domain knowledge of object-based domains to design three types of event-based convolutional layers to direct exploration. These layers rely on Gaussian dropouts and are inserted between the layers of the deep neural network model to help facilitate variational Thompson sampling. We empirically show the effectiveness of EVaDE-equipped Simulated Policy Learning (EVaDE-SimPLe) on the 100K Atari game suite.