Hyperdimensional Decoding of Spiking Neural Networks

2025-11-11

AI Summary

Addressing the challenge of simultaneously achieving high accuracy, robustness, low latency, and energy efficiency in spiking neural network (SNN) decoding, this paper proposes the first SNN decoding framework integrated with hyperdimensional computing (HDC). The method employs binary hypervectors for efficient encoding and similarity-based matching, transcending conventional rate- or latency-coding paradigms and enabling zero-shot category recognition. By synergistically combining HDC’s semantic robustness with SNNs’ event-driven computation, it significantly reduces inference-time computational overhead. Evaluated on DvsGesture and SL-Animals-DVS datasets, the framework achieves 1.24–3.67× lower energy consumption than baseline methods, improved classification accuracy, reduced latency, and perfect (100%) recognition of unseen categories. This work establishes a novel paradigm for neuromorphic decoding that is both ultra-low-power and highly generalizable.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Hardware-aware ML
• Computer Vision: Learning & Optimization for CV

Application Category

• Search and Retrieval-Augmented AI: Web query analysis, representation and understanding
• Graph Algorithms and Modeling for the Web: Graph neural networks and deep learning approaches for Web-related graphs
• Economics, Online Markets and Human Computation: Cost models of using LLMs in production systems

Abstract

This work presents a novel spiking neural network (SNN) decoding method, combining SNNs with Hyperdimensional computing (HDC). The goal is to create a decoding method with high accuracy, high noise robustness, low latency and low energy usage. Compared to analogous architectures decoded with existing approaches, the presented SNN-HDC model attains generally better classification accuracy, lower classification latency and lower estimated energy consumption on multiple test cases from literature. The SNN-HDC achieved estimated energy consumption reductions ranging from 1.24x to 3.67x on the DvsGesture dataset and from 1.38x to 2.27x on the SL-Animals-DVS dataset. The presented decoding method can also efficiently identify unknown classes it has not been trained on. In the DvsGesture dataset the SNN-HDC model can identify 100% of samples from an unseen/untrained class. Given the numerous benefits shown and discussed in this paper, this decoding method represents a very compelling alternative to both rate and latency decoding.

Problem

• Develops a spiking neural network decoding method with hyperdimensional computing
• Achieves high accuracy, noise robustness, low latency and energy consumption
• Enables efficient identification of unknown classes not seen during training

Innovation

• Combining spiking neural networks with hyperdimensional computing
• Achieving lower energy consumption and higher accuracy
• Identifying unknown classes not seen during training