AI RESEARCH
DualFlexKAN: Dual-stage Kolmogorov-Arnold Networks with Independent Function Control
arXiv CS.LG
•
ArXi:2603.08583v1 Announce Type: new 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. Kolmogoro-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.