Recent Advances in Scalable Energy-Efficient and Trustworthy Spiking Neural networks: from Algorithms to Technology

TL;DR

This paper reviews recent algorithmic, architectural, and hardware innovations in scalable, energy-efficient, and trustworthy spiking neural networks, highlighting their potential for advanced signal processing applications.

Contribution

It provides a comprehensive overview of recent advances in algorithms, hardware, and co-design strategies for scalable and trustworthy SNNs, emphasizing the integration of computation within memory and sensors.

Findings

Recent algorithms enable efficient training of low-latency SNNs

Hardware designs leverage in-memory and sensor integration for efficiency

Co-design approaches balance energy, latency, and accuracy

Abstract

Neuromorphic computing and, in particular, spiking neural networks (SNNs) have become an attractive alternative to deep neural networks for a broad range of signal processing applications, processing static and/or temporal inputs from different sensory modalities, including audio and vision sensors. In this paper, we start with a description of recent advances in algorithmic and optimization innovations to efficiently train and scale low-latency, and energy-efficient spiking neural networks (SNNs) for complex machine learning applications. We then discuss the recent efforts in algorithm-architecture co-design that explores the inherent trade-offs between achieving high energy-efficiency and low latency while still providing high accuracy and trustworthiness. We then describe the underlying hardware that has been developed to leverage such algorithmic innovations in an efficient way. In particular, we describe a hybrid method to integrate significant portions of the model's computation within both memory components as well as the sensor itself. Finally, we discuss the potential path forward for research in building deployable SNN systems identifying key challenges in the algorithm-hardware-application co-design space with an emphasis on trustworthiness.