Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

Energy-Efficient Neuromorphic Computing for Edge AI: A Framework with Adaptive Spiking Neural Networks and Hardware-Aware Optimization

Olaf Yunus Laitinen Imanov, Derya Umut Kulali, Taner Yilmaz, Duygu Erisken, Rana Irem Turhan

TL;DR

NeuEdge tackles the challenge of energy-efficient edge AI by integrating adaptive spiking neural networks with hardware-aware optimization. The framework combines a hybrid temporal encoder, hardware-aware co-optimization, an adaptive threshold mechanism, and a training procedure that respects on-chip constraints, achieving 91–96% accuracy and 847 GOp/s/W across vision and audio tasks with a 2.3 ms latency. Key contributions include a 4.7× spike reduction via hybrid encoding, 89% hardware utilization on Loihi 2, and up to 312× energy savings over conventional GPUs, demonstrated on multiple neuromorphic platforms and edge processors. The work significantly advances practical neuromorphic edge AI by delivering holistic, end-to-end optimization from encoding to deployment and real-world workload validation.

Abstract

Edge AI applications increasingly require ultra-low-power, low-latency inference. Neuromorphic computing based on event-driven spiking neural networks (SNNs) offers an attractive path, but practical deployment on resource-constrained devices is limited by training difficulty, hardware-mapping overheads, and sensitivity to temporal dynamics. We present NeuEdge, a framework that combines adaptive SNN models with hardware-aware optimization for edge deployment. NeuEdge uses a temporal coding scheme that blends rate and spike-timing patterns to reduce spike activity while preserving accuracy, and a hardware-aware training procedure that co-optimizes network structure and on-chip placement to improve utilization on neuromorphic processors. An adaptive threshold mechanism adjusts neuron excitability from input statistics, reducing energy consumption without degrading performance. Across standard vision and audio benchmarks, NeuEdge achieves 91-96% accuracy with up to 2.3 ms inference latency on edge hardware and an estimated 847 GOp/s/W energy efficiency. A case study on an autonomous-drone workload shows up to 312x energy savings relative to conventional deep neural networks while maintaining real-time operation.

Energy-Efficient Neuromorphic Computing for Edge AI: A Framework with Adaptive Spiking Neural Networks and Hardware-Aware Optimization

TL;DR

NeuEdge tackles the challenge of energy-efficient edge AI by integrating adaptive spiking neural networks with hardware-aware optimization. The framework combines a hybrid temporal encoder, hardware-aware co-optimization, an adaptive threshold mechanism, and a training procedure that respects on-chip constraints, achieving 91–96% accuracy and 847 GOp/s/W across vision and audio tasks with a 2.3 ms latency. Key contributions include a 4.7× spike reduction via hybrid encoding, 89% hardware utilization on Loihi 2, and up to 312× energy savings over conventional GPUs, demonstrated on multiple neuromorphic platforms and edge processors. The work significantly advances practical neuromorphic edge AI by delivering holistic, end-to-end optimization from encoding to deployment and real-world workload validation.

Abstract

Edge AI applications increasingly require ultra-low-power, low-latency inference. Neuromorphic computing based on event-driven spiking neural networks (SNNs) offers an attractive path, but practical deployment on resource-constrained devices is limited by training difficulty, hardware-mapping overheads, and sensitivity to temporal dynamics. We present NeuEdge, a framework that combines adaptive SNN models with hardware-aware optimization for edge deployment. NeuEdge uses a temporal coding scheme that blends rate and spike-timing patterns to reduce spike activity while preserving accuracy, and a hardware-aware training procedure that co-optimizes network structure and on-chip placement to improve utilization on neuromorphic processors. An adaptive threshold mechanism adjusts neuron excitability from input statistics, reducing energy consumption without degrading performance. Across standard vision and audio benchmarks, NeuEdge achieves 91-96% accuracy with up to 2.3 ms inference latency on edge hardware and an estimated 847 GOp/s/W energy efficiency. A case study on an autonomous-drone workload shows up to 312x energy savings relative to conventional deep neural networks while maintaining real-time operation.
Paper Structure (32 sections, 11 equations, 4 figures, 4 tables, 1 algorithm)

This paper contains 32 sections, 11 equations, 4 figures, 4 tables, 1 algorithm.

Table of Contents

  1. Introduction
  2. Related Work
  3. Spiking Neural Networks
  4. Neuromorphic Hardware
  5. Edge AI and Efficient Neural Networks
  6. Neuromorphic Edge AI
  7. Research Gaps
  8. Background
  9. Spiking Neuron Models
  10. Spike Encoding
  11. Energy Model
  12. NeuEdge Framework
  13. Architecture Overview
  14. Hybrid Temporal Coding
  15. Hardware-Aware Co-Optimization
  16. ...and 17 more sections

Figures (4)

  • Figure 1: NeuEdge framework architecture integrating hybrid temporal encoding, hardware-aware co-optimization, adaptive training, and runtime optimization for energy-efficient edge deployment.
  • Figure 2: Resource utilization on Loihi 2 comparing NeuEdge with naive mapping (cores, synapses, and inter-core traffic).
  • Figure 3: Energy breakdown per inference on Loihi 2 across computation and communication components.
  • Figure 4: Spike counts per inference for different encoding/training approaches (CIFAR-10).