在强化学习框架下缓解二维矩形装箱问题的维度问题
Mitigating Dimensionality in 2D Rectangle Packing Problem under Reinforcement Learning Schema
AI总结:
本文将强化学习应用于二维矩形装箱问题,通过简化状态与动作空间表示并结合UNet架构与PPO算法,实现了与MaxRect启发式算法相当的模型,且在泛化至非矩形装箱和复杂约束上具巨大潜力。
摘要:
本文探讨了将强化学习(RL)应用于二维矩形装箱问题。我们提出了一种状态和动作空间的简化表示,使我们能够实现高粒度。利用UNet架构和近端策略优化(PPO),我们实现了一个与MaxRect启发式算法相当的模型。然而,我们的方法在泛化到非矩形装箱问题和复杂约束方面具有巨大潜力。
英文摘要:
This paper explores the application of Reinforcement Learning (RL) to the two-dimensional rectangular packing problem. We propose a reduced representation of the state and action spaces that allow us for high granularity. Leveraging UNet architecture and Proximal Policy Optimization (PPO), we achieved a model that is comparable to the MaxRect heuristic. However, our approach has great potential to be generalized to nonrectangular packing problems and complex constraints.
机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。