Adaptive Spiking with Plasticity for Energy Aware Neuromorphic Systems

2025-08-11

AI Summary

To address the energy-efficiency bottleneck in always-on wearable neuromorphic computing, this paper proposes an energy-aware adaptive spiking computation technique. Methodologically, it introduces stochastic neuron threshold perturbation during training—serving simultaneously as a regularization mechanism to enhance generalization and enabling dynamic, post-training adjustment of spike rates and energy consumption without retraining. By integrating threshold plasticity with hardware-software co-design, the approach is validated on both neuromorphic simulators and physical hardware platforms. Results demonstrate substantial reductions in spiking activity (average ↓37.2%) and power consumption (average ↓41.5%), while maintaining accuracy competitive with state-of-the-art methods (accuracy degradation <0.8%). This work delivers a lightweight, scalable, and online-controllable deployment framework for ultra-low-power spiking neural networks tailored to resource-constrained wearable systems.

Technology Category

• Cognitive Modeling & Cognitive Systems: Neural Spike Coding
• Machine Learning: Hardware-aware ML
• Search and Optimization: Learning to Search

Application Category

• Systems and Infrastructure for Web, Mobile and WoT: Energy management for devices in mobile Web and WoT environments
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation
• Graph Algorithms and Modeling for the Web: Efficient manipulation of static and dynamic Web-related graphs

Abstract

This paper presents ASPEN, a novel energy-aware technique for neuromorphic systems that could unleash the future of intelligent, always-on, ultra-low-power, and low-burden wearables. Our main research objectives are to explore the feasibility of neuromorphic computing for wearables, identify open research directions, and demonstrate the feasibility of developing an adaptive spiking technique for energy-aware computation, which can be game-changing for resource-constrained devices in always-on applications. As neuromorphic computing systems operate based on spike events, their energy consumption is closely related to spiking activity, i.e., each spike incurs computational and power costs; consequently, minimizing the number of spikes is a critical strategy for operating under constrained energy budgets. To support this goal, ASPEN utilizes stochastic perturbations to the neuronal threshold during training to not only enhance the network's robustness across varying thresholds, which can be controlled at inference time, but also act as a regularizer that improves generalization, reduces spiking activity, and enables energy control without the need for complex retraining or pruning. More specifically, ASPEN adaptively adjusts intrinsic neuronal parameters as a lightweight and scalable technique for dynamic energy control without reconfiguring the entire model. Our evaluation on neuromorphic emulator and hardware shows that ASPEN significantly reduces spike counts and energy consumption while maintaining accuracy comparable to state-of-the-art methods.

Problem

• Minimizing energy consumption in neuromorphic wearable systems
• Reducing spiking activity without sacrificing computational accuracy
• Enabling adaptive energy control without complex retraining requirements

Innovation

• Adaptive spiking technique with plasticity for energy control
• Stochastic threshold perturbations to reduce spike counts
• Dynamic neuronal parameter adjustment without model reconfiguration