Model Releases
Learning with Volterra Neural Networks: A System Theoretic Perspective
arXiv:2609.01928v1 Announce Type: new Abstract: Higher-order interaction components are important for signal, image, and video modeling, but explicit high-order operators often suffer from rapidly inc
arXiv:2609.01928v1 Announce Type: new Abstract: Higher-order interaction components are important for signal, image, and video modeling, but explicit high-order operators often suffer from rapidly increasing parameter and computational costs. This paper presents kVNN, a learnable kernelized Volterra Neural operator for compact higher-order filtering. The motivation is to use kernelization to improve the efficiency of Volterra-type neural operators while providing a structured interpretation of their higher-order components. The proposed formulation combines the order-wise structure of Volterra filtering with learnable polynomial-kernel atoms, allowing different interaction orders to be represented by separate learnable centers and coefficients. This order-decoupled representation avoids explicit high-order tensor parameterization and can be implemented as a CNN-compatible layer. Experiments on representative vision tasks show that kVNN achieves a favorable accuracy--efficiency trade-off.
Related
- Learning Explicit Physical Parameter Control and Benchmarking for Video Generation
- FedStain: Modeling Higher-Order Stain Statistics for Federated Domain Generalization in Computational Pathology
- GroupKAN: Efficient Kolmogorov-Arnold Networks via Grouped Spline Modeling
- A Unified Framework for the Mechanics of Information in Convolutional Neural Network Image Space
Source: arXiv cs.CV | 2026-09-03