Perspective: an optoelectronic future for heterogeneous, dendritic computing

Luis El Srouji; Mahmoud Abdelghany; Hari Rakul Ambethkar; Yun-Jhu Lee; Mehmet Berkay On; S J Ben Yoo

doi:10.3389/fnins.2024.1394271

Perspective: an optoelectronic future for heterogeneous, dendritic computing

Front Neurosci. 2024 Apr 17:18:1394271. doi: 10.3389/fnins.2024.1394271. eCollection 2024.

Authors

Luis El Srouji¹, Mahmoud Abdelghany¹, Hari Rakul Ambethkar¹, Yun-Jhu Lee¹, Mehmet Berkay On¹, S J Ben Yoo¹

Affiliation

¹ Department of Electrical and Computer Engineering, University of California, Davis, Davis, CA, United States.

Abstract

With the increasing number of applications reliant on large neural network models, the pursuit of more suitable computing architectures is becoming increasingly relevant. Progress toward co-integrated silicon photonic and CMOS circuits provides new opportunities for computing architectures with high bandwidth optical networks and high-speed computing. In this paper, we discuss trends in neuromorphic computing architecture and outline an optoelectronic future for heterogeneous, dendritic neuromorphic computing.

Keywords: analog computing; dendritic computing; heterogeneous computing; neuromorphic computing; silicon photonic computing.

Grants and funding

The author(s) declare financial support was received for the research, authorship, and/or publication of this article. This research was based upon work supported in part by the Office of the Director of National Intelligence (ODNI), Intelligence Advanced Research Projects Activity (IARPA), via [2021-21090200004]. The views and conclusions contained herein are those of the authors and should not be interpreted as necessarily representing the official policies, either expressed or implied, of ODNI, IARPA, or the U.S. Government. The U.S. Government was authorized to reproduce and distribute reprints for governmental purposes notwithstanding any copyright annotation therein.