Can Agents Learn by Analogy? An Inferable Model for PAC Reinforcement Learning
This addresses the challenge of sample and computational inefficiency in reinforcement learning for agents, though it appears incremental as it builds on existing model-based approaches with a novel inference mechanism.
The paper tackles the problem of model-based reinforcement learning by proposing Greedy Inference Model (GIM), which infers unknown dynamics from known ones using spectral properties, enabling 'learning by analogy' and reducing computational cost. Experimental results show GIM is more computationally efficient than state-of-the-art methods, with sample complexity improvements under mild conditions.
Model-based reinforcement learning algorithms make decisions by building and utilizing a model of the environment. However, none of the existing algorithms attempts to infer the dynamics of any state-action pair from known state-action pairs before meeting it for sufficient times. We propose a new model-based method called Greedy Inference Model (GIM) that infers the unknown dynamics from known dynamics based on the internal spectral properties of the environment. In other words, GIM can "learn by analogy". We further introduce a new exploration strategy which ensures that the agent rapidly and evenly visits unknown state-action pairs. GIM is much more computationally efficient than state-of-the-art model-based algorithms, as the number of dynamic programming operations is independent of the environment size. Lower sample complexity could also be achieved under mild conditions compared against methods without inferring. Experimental results demonstrate the effectiveness and efficiency of GIM in a variety of real-world tasks.