Human-Inspired Framework to Accelerate Reinforcement Learning
This addresses sample inefficiency in reinforcement learning for data science applications, but it appears incremental as it builds on existing transfer learning approaches without a major paradigm shift.
The paper tackles the problem of sample inefficiency in reinforcement learning by introducing a human-inspired framework that exposes agents to progressively complex tasks, achieving improved sample efficiency in challenging main tasks like optimal control problems with constraints.
Reinforcement learning (RL) is crucial for data science decision-making but suffers from sample inefficiency, particularly in real-world scenarios with costly physical interactions. This paper introduces a novel human-inspired framework to enhance RL algorithm sample efficiency. It achieves this by initially exposing the learning agent to simpler tasks that progressively increase in complexity, ultimately leading to the main task. This method requires no pre-training and involves learning simpler tasks for just one iteration. The resulting knowledge can facilitate various transfer learning approaches, such as value and policy transfer, without increasing computational complexity. It can be applied across different goals, environments, and RL algorithms, including value-based, policy-based, tabular, and deep RL methods. Experimental evaluations demonstrate the framework's effectiveness in enhancing sample efficiency, especially in challenging main tasks, demonstrated through both a simple Random Walk and more complex optimal control problems with constraints.