Can KAN Work? Exploring the Potential of Kolmogorov-Arnold Networks in Computer Vision

TL;DR

This paper investigates the application of Kolmogorov-Arnold Networks (KANs) in computer vision, assessing their performance, robustness issues, and proposing regularization techniques to improve their stability and generalization in image tasks.

Contribution

It is the first comprehensive analysis of KANs in computer vision, introducing regularization methods to enhance their robustness and applicability.

Findings

KANs have strong fitting capabilities but are sensitive to noise.

Regularization methods improve KAN stability and generalization.

KANs show potential in complex visual data tasks.

Abstract

Kolmogorov-Arnold Networks(KANs), as a theoretically efficient neural network architecture, have garnered attention for their potential in capturing complex patterns. However, their application in computer vision remains relatively unexplored. This study first analyzes the potential of KAN in computer vision tasks, evaluating the performance of KAN and its convolutional variants in image classification and semantic segmentation. The focus is placed on examining their characteristics across varying data scales and noise levels. Results indicate that while KAN exhibits stronger fitting capabilities, it is highly sensitive to noise, limiting its robustness. To address this challenge, we propose a smoothness regularization method and introduce a Segment Deactivation technique. Both approaches enhance KAN's stability and generalization, demonstrating its potential in handling complex visual data tasks.