在神经形态芯片上实现低功耗持续学习，避免遗忘且速度提升百倍。
Online Continual Learning on Intel Loihi 2 via a Co-designed Spiking Neural Network
- 自归一化脉冲学习规则+事件驱动状态机，支持芯片原生持续学习
- 比传统边缘GPU快113倍、省电6600倍，准确率媲美带重放的模型
- 适合低功耗实时场景，如智能传感器、可穿戴设备
边缘设备上的AI系统需在严格功耗限制下实现在线持续学习——在非平稳数据流中适应新类别，同时避免灾难性遗忘。我们提出CLP-SNN，一种结合自归一化局部学习规则与脉冲驱动神经状态机的脉冲神经网络，部署于英特尔Loihi 2神经形态处理器。在OpenLORIS少样本实验中，CLP-SNN在无重放情况下达到与基于重放的模型相当的准确率。在Loihi 2上，其延迟仅为0.33毫秒（对比边缘GPU的37.3毫秒），能耗仅0.05毫焦（对比333毫焦），分别降低113倍和6600倍。该性能提升来自算法效率（约14.5倍延迟、22.6倍能耗优化）与硬件协同设计（约7.8倍延迟、295倍能耗优化），充分利用事件驱动学习与稀疏分级脉冲通信。结果表明，协同设计的类脑算法与神经形态硬件可突破传统边缘AI的精度-效率权衡瓶颈。
原文摘要 · Abstract (English)
AI systems on edge devices require online continual learning -- adapting to non-stationary streams and unfamiliar classes without catastrophic forgetting -- under strict power constraints. We present CLP-SNN, a spiking neural network with a self-normalizing local learning rule and a spike-driven neural state machine for autonomous on-chip learning, implemented on Intel's Loihi 2 neuromorphic processor. On OpenLORIS few-shot experiments, CLP-SNN matches replay-based accuracy rehearsal-free. On Loihi 2, CLP-SNN achieves 113x lower latency (0.33 ms vs. 37.3 ms) and 6,600x lower energy (0.05 mJ vs. 333 mJ) than the strongest edge-GPU baseline. This gain decomposes into algorithmic efficiency (~14.5x latency, ~22.6x energy on the same GPU) and neuromorphic hardware co-design (~7.8x latency, ~295x energy) exploiting event-driven learning and sparse graded-spike communication. We show that co-designed brain-inspired algorithms and neuromorphic hardware can break traditional accuracy-efficiency trade-offs in edge AI.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。