Chen, Demeng, Tran, Vince, Genov, Roman, Rahimi Azghadi, Mostafa, Ahmadi, Majid, and Amirsoleimani, Amirali (2025) MAKAN: Memristive Accelerated Kolmugruv-Arnold Networks. In: Proceedings of the Midwest Symposium on Circuits and Systems. pp. 478-482. From: MWSCAS 2025: IEEE 68th International Midwest Symposium on Circuits and Systems, 10-13 August 2025, Lansing, MI, USA.

PDF (Published Version) - Published Version Restricted to Repository staff only

 

2

Abstract

Traditional computing architectures struggle to keep up with modern neural networks due to energy consumption and processing speed limitations. To address these challenges, we propose MAKAN, a framework that integrates Kolmogorov-Arnold Networks (KANs) with the BITLITE memristor-based computing platform. KANs, which use learnable activation functions on edges instead of fixed functions on nodes, offer superior flexibility and accuracy, but their reliance on spline-based functions limits scalability. MAKAN overcomes this by using hat functions for piecewise linear (PWL) function approximation, efficiently implemented on BITLITE's bit-wise memristor crossbars for matrix-vector multiplication (MVM). By employing machine learning to reconstruct KAN's spline functions with PWL on memristor circuits, MAKAN significantly reduces computational overhead while preserving accuracy and interpretability. This integration results in notable improvements in energy efficiency and processing speed, making advanced neural networks more scalable and practical for real-world applications, paving the way for future advancements in energy-efficient machine learning systems.

Actions (Repository Staff Only)

Item Control Page