SINDy-RL: Interpretable and Efficient Model-Based Reinforcement Learning
CoRR(2024)
摘要
Deep reinforcement learning (DRL) has shown significant promise for
uncovering sophisticated control policies that interact in environments with
complicated dynamics, such as stabilizing the magnetohydrodynamics of a tokamak
fusion reactor or minimizing the drag force exerted on an object in a fluid
flow. However, these algorithms require an abundance of training examples and
may become prohibitively expensive for many applications. In addition, the
reliance on deep neural networks often results in an uninterpretable, black-box
policy that may be too computationally expensive to use with certain embedded
systems. Recent advances in sparse dictionary learning, such as the sparse
identification of nonlinear dynamics (SINDy), have shown promise for creating
efficient and interpretable data-driven models in the low-data regime. In this
work we introduce SINDy-RL, a unifying framework for combining SINDy and DRL to
create efficient, interpretable, and trustworthy representations of the
dynamics model, reward function, and control policy. We demonstrate the
effectiveness of our approaches on benchmark control environments and
challenging fluids problems. SINDy-RL achieves comparable performance to
state-of-the-art DRL algorithms using significantly fewer interactions in the
environment and results in an interpretable control policy orders of magnitude
smaller than a deep neural network policy.
更多查看译文
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要