DualFlexKAN: Dual-stage Kolmogorov-Arnold Networks with Independent Function Control

TL;DR

DualFlexKAN introduces a dual-stage, flexible neural network architecture that independently controls input transformations and output activations, enabling more efficient and expressive function approximation with fewer parameters.

Contribution

It proposes a novel dual-stage architecture that supports diverse basis functions and regularization, improving over standard KANs in efficiency and flexibility.

Findings

Outperforms MLPs and KANs in accuracy and convergence.

Achieves 10-100x fewer parameters than standard KANs.

Excels in regression, physics-informed tasks, and function approximation.

Abstract

Multi-Layer Perceptrons (MLPs) rely on pre-defined, fixed activation functions, imposing a static inductive bias that forces the network to approximate complex topologies solely through increased depth and width. Kolmogorov-Arnold Networks (KANs) address this limitation through edge-centric learnable functions, yet their formulation suffers from quadratic parameter scaling and architectural rigidity that hinders the effective integration of standard regularization techniques. This paper introduces the DualFlexKAN (DFKAN), a flexible architecture featuring a dual-stage mechanism that independently controls pre-linear input transformations and post-linear output activations. This decoupling enables hybrid networks that optimize the trade-off between expressiveness and computational cost. Unlike standard formulations, DFKAN supports diverse basis function families, including orthogonal polynomials, B-splines, and radial basis functions, integrated with configurable regularization strategies that stabilize training dynamics. Comprehensive evaluations across regression benchmarks, physics-informed tasks, and function approximation demonstrate that DFKAN outperforms both MLPs and conventional KANs in accuracy, convergence speed, and gradient fidelity. The proposed hybrid configurations achieve superior performance with one to two orders of magnitude fewer parameters than standard KANs, effectively mitigating the parameter explosion problem while preserving KAN-style expressiveness. DFKAN provides a principled, scalable framework for incorporating adaptive non-linearities, proving particularly advantageous for data-efficient learning and interpretable function discovery in scientific applications.