Bridging Synapses: A Comparative Review of Machine Learning Algorithms in Memristor Technology

Authors

  • Prashant Dhakal Department of Industrial and Systems Engineering, Texas A&M University, College Station, TX, USA
  • Xiaofei Wu Department of Industrial and Systems Engineering, Texas A&M University, College Station, TX, USA
  • Jae Gwang Kim Department of Material Science and Engineering, Texas A&M University, College Station, TX, USA
  • Aolin Hou Department of Material Science and Engineering, Texas A&M University, College Station, TX, USA
  • Shiren Wang Department of Industrial and Systems Engineering, Texas A&M University, College Station, TX, USA

Keywords:

Neuromorphic computing, memristors, AI, neuromorphic engineering, machine learning, AI hardware, algorithm

Abstract

The integration of memristor-based nanodevices into machine learning systems has garnered significant attention due to their rapid, low-energy, and nonvolatile switching capabilities. Memristors, with their unique ability to retain an internal resistance state based on voltage and current history, excel in executing core operations like vector matrix multiplication, addressing the von Neumann bottleneck. This review explores the synergy between memristors and machine learning, highlighting their potential to enhance neuromorphic computing through energy-efficient and highly parallel architectures. The paper examines current research, practical implementations, and emerging challenges, providing a comprehensive analysis of the future directions in this interdisciplinary field.

Downloads

Download data is not yet available.

Downloads

Published

2024-06-21

How to Cite

Dhakal, P., Wu, X., Kim, J. G., Hou, A., & Wang, S. (2024). Bridging Synapses: A Comparative Review of Machine Learning Algorithms in Memristor Technology. Journal of Neuromorphic Intelligence, 1(1). Retrieved from https://sci-access.org/index.php/nsej/article/view/7

Issue

Vol. 1 No. 1 (2024): Launch Issue

Section

Articles