Work

Kolmogorov-Arnold Network for Online Reinforcement Learning

Year: 2024

Type: preprint

Abstract: 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.

Source: arXiv (Cornell University)

DOI: 10.48550/arxiv.2408.04841

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

Institutions: University of Tsukuba, Universidade Federal de Santa Maria, Universidad Tecnológica de Uruguay

Language: English

Topic: Reinforcement Learning in Robotics

Subfield: Artificial Intelligence

Field: Computer Science

Domain: Physical Sciences

Open Access status: green

Funders: Ministry of Education, Culture, Sports, Science and Technology, University of Tsukuba

OpenAlex ID: W4402952650

Cites:

Cited by:

Related to: 10