arXiv 2408.048412024-08-31cs.LGcs.AI

用于在线强化学习的Kolmogorov-Arnold网络

Kolmogorov-Arnold Network for Online Reinforcement Learning

University of Tsukuba（筑波大学）；Federal University of Santa Maria（圣玛丽亚联邦大学）；Technological University of Uruguay（乌拉圭科技大学）

Victor Augusto Kich, Jair Augusto Bottega, Raul Steinmetz, Ricardo Bedin Grando, Ayano Yorozu, Akihisa Ohya

备注：: Paper accepted at 24th International Conference on Control, Automation and Systems (ICCAS)

阅读 PDF

arXiv 页面 ↗HTML 原文 ↗

AI总结：

本文研究将KANs作为PPO算法中的函数逼近器，并在DeepMind Control Proprio Robotics基准上验证其能以更少参数达到与MLP相当的性能。

摘要：

Kolmogorov-Arnold网络（KANs）已展现出作为神经网络中多层感知器（MLPs）替代方案的潜力，能够以更少的参数和更低的内存占用提供通用函数逼近能力。在本文中，我们探索了将KANs用作近端策略优化（PPO）算法中的函数逼近器。我们通过使用DeepMind Control Proprio Robotics基准，将该方法与基于MLP的原始PPO进行性能比较来评估这一方案。我们的结果表明，基于KAN的强化学习算法能够达到与其基于MLP的对应方法相当的性能，且通常使用更少的参数。这些发现表明，KANs可能为强化学习模型提供一种更高效的选择。

英文摘要：

Kolmogorov-Arnold Networks (KANs) have shown potential as an alternative to Multi-Layer Perceptrons (MLPs) in neural networks, providing universal function approximation with fewer parameters and reduced memory usage. In this paper, we explore the use of KANs as function approximators within the Proximal Policy Optimization (PPO) algorithm. We evaluate this approach by comparing its performance to the original MLP-based PPO using the DeepMind Control Proprio Robotics benchmark. Our results indicate that the KAN-based reinforcement learning algorithm can achieve comparable performance to its MLP-based counterpart, often with fewer parameters. These findings suggest that KANs may offer a more efficient option for reinforcement learning models.

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。