Neural scaling laws and evolution of learnable activation functions of Kolmogorov-Arnold networks

arXiv:2610.00985 · cs.LG · Submitted 2026-10-01 · Read on arXiv

cs.LG

Submitted: 2026-10-01

Updated: 2026-10-01

Code: https://github.com/hoangthangta/BSRBF

Terminology

• Activation Function — 1× in this paper · explained in 1 episode
• Scaling Laws — 1× in this paper · explained in 4 episodes

Sources

• KAN: Kolmogorov-Arnold Networks
• A Practitioner's Guide to Kolmogorov-Arnold Networks
• PRKAN: Parameter-Reduced Kolmogorov-Arnold Networks
• Fashion-MNIST: a Novel Image Dataset for Benchmarking Machine Learning Algorithms
• Neural Scaling Laws for Deep Regression
• Scaling Laws for Neural Language Models
• Training Compute-Optimal Large Language Models
• BSRBF-KAN: A combination of B-splines and Radial Basis Functions in Kolmogorov-Arnold Networks
• Exploring the Potential of Polynomial Basis Functions in Kolmogorov-Arnold Networks: A Comparative Study of Different Groups of Polynomials
• Broken Neural Scaling Laws
• A Solvable Model of Neural Scaling Laws
• An Analysis of Honeypots and their Impact as a Cyber Deception Tactic

Related papers

• Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
• AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
• Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
• Convergence issues in Relational Concept Analysis based on AOC-posets
• Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
• Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks