Neuromorphic Sequential Arena: A Benchmark for Neuromorphic Temporal Processing

TL;DR

The paper introduces the Neuromorphic Sequential Arena (NSA), a benchmark for evaluating neuromorphic temporal processing in SNNs across diverse real-world tasks, facilitating systematic comparison and advancement in the field.

Contribution

It presents NSA as a comprehensive, standardized benchmark for neuromorphic temporal processing, including diverse tasks and baseline evaluations of recent models.

Findings

NSA enables systematic tracking of progress in neuromorphic temporal processing.

Recent SNN models vary significantly in performance, efficiency, and training speed.

Efficient SNN designs are needed for high performance across diverse temporal tasks.

Abstract

Temporal processing is vital for extracting meaningful information from time-varying signals. Recent advancements in Spiking Neural Networks (SNNs) have shown immense promise in efficiently processing these signals. However, progress in this field has been impeded by the lack of effective and standardized benchmarks, which complicates the consistent measurement of technological advancements and limits the practical applicability of SNNs. To bridge this gap, we introduce the Neuromorphic Sequential Arena (NSA), a comprehensive benchmark that offers an effective, versatile, and application-oriented evaluation framework for neuromorphic temporal processing. The NSA includes seven real-world temporal processing tasks from a diverse range of application scenarios, each capturing rich temporal dynamics across multiple timescales. Utilizing NSA, we conduct extensive comparisons of recently introduced spiking neuron models and neural architectures, presenting comprehensive baselines in terms of task performance, training speed, memory usage, and energy efficiency. Our findings emphasize an urgent need for efficient SNN designs that can consistently deliver high performance across tasks with varying temporal complexities while maintaining low computational costs. NSA enables systematic tracking of advancements in neuromorphic algorithm research and paves the way for developing effective and efficient neuromorphic temporal processing systems.