NeuroMorse: A Temporally Structured Dataset For Neuromorphic Computing

2025-02-28

AI Summary

Existing neuromorphic benchmarks predominantly emphasize spatial features, neglecting the critical temporal dynamics essential for sequence tasks, thereby inadequately evaluating brain-inspired systems’ capacity to learn spatiotemporal joint representations. To address this gap, we introduce NeuroMorse—the first time-structured benchmark explicitly designed for spike-timing modeling: it maps high-frequency English words to precisely timed Morse code spike sequences, encoded solely via dot/dash dual-channel pulses that capture multiscale temporal structure. NeuroMorse supports asynchronous event-sequence modeling and spiking neural network (SNN) evaluation. Experiments reveal severe performance limitations of linear classifiers on training data and poor generalization of conventional methods to unseen test keywords, whereas brain-inspired models demonstrate marked advantages in temporal decoding. This work fills a critical void in realistic spatiotemporal benchmarking and establishes a standardized evaluation platform for neuromorphic temporal learning.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Multimodal Learning
• Planning, Routing, and Scheduling: Optimization of Spatio-temporal Systems

Application Category

• Search and Retrieval-Augmented AI: Web evaluation methodologies and metrics
• Graph Algorithms and Modeling for the Web: Algorithms and analysis for heterogeneous, signed, attributed, multi-relational, temporal, higher-order, and annotated Web-related graphs
• Systems and Infrastructure for Web, Mobile and WoT: Web performance, measurement, and characterization

Abstract

Neuromorphic engineering aims to advance computing by mimicking the brain's efficient processing, where data is encoded as asynchronous temporal events. This eliminates the need for a synchronisation clock and minimises power consumption when no data is present. However, many benchmarks for neuromorphic algorithms primarily focus on spatial features, neglecting the temporal dynamics that are inherent to most sequence-based tasks. This gap may lead to evaluations that fail to fully capture the unique strengths and characteristics of neuromorphic systems. In this paper, we present NeuroMorse, a temporally structured dataset designed for benchmarking neuromorphic learning systems. NeuroMorse converts the top 50 words in the English language into temporal Morse code spike sequences. Despite using only two input spike channels for Morse dots and dashes, complex information is encoded through temporal patterns in the data. The proposed benchmark contains feature hierarchy at multiple temporal scales that test the capacity of neuromorphic algorithms to decompose input patterns into spatial and temporal hierarchies. We demonstrate that our training set is challenging to categorise using a linear classifier and that identifying keywords in the test set is difficult using conventional methods. The NeuroMorse dataset is available at Zenodo, with our accompanying code on GitHub at https://github.com/Ben-E-Walters/NeuroMorse.

Problem

• Addresses lack of temporal dynamics in neuromorphic benchmarks
• Introduces NeuroMorse dataset for temporal pattern benchmarking
• Tests neuromorphic algorithms on multi-scale temporal hierarchies

Innovation

• Temporal Morse code spike sequences encoding
• Multi-scale temporal feature hierarchy testing
• Two-channel input for complex temporal patterns