Can KAN Work? Exploring the Potential of Kolmogorov-Arnold Networks in Computer Vision
This work addresses the robustness problem of KANs in computer vision for researchers and practitioners, but it is incremental as it builds on existing KAN architecture with modifications.
This study tackled the application of Kolmogorov-Arnold Networks (KANs) to computer vision tasks like image classification and semantic segmentation, finding that while KANs have strong fitting capabilities, they are highly sensitive to noise, and proposed smoothness regularization and Segment Deactivation techniques to enhance stability and generalization.
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.