Online Continual Learning on Intel Loihi 2 via a Co-designed Spiking Neural Network
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2025
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866910193315479552 |
|---|---|
| author | Hajizada, Elvin Rager, Danielle Shea, Timothy Campos-Macias, Leobardo Wild, Andreas Hüllermeier, Eyke Sandamirskaya, Yulia Davies, Mike |
| author_facet | Hajizada, Elvin Rager, Danielle Shea, Timothy Campos-Macias, Leobardo Wild, Andreas Hüllermeier, Eyke Sandamirskaya, Yulia Davies, Mike |
| contents | 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. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2511_01553 |
| institution | arXiv |
| publishDate | 2025 |
| record_format | arxiv |
| spellingShingle | Online Continual Learning on Intel Loihi 2 via a Co-designed Spiking Neural Network Hajizada, Elvin Rager, Danielle Shea, Timothy Campos-Macias, Leobardo Wild, Andreas Hüllermeier, Eyke Sandamirskaya, Yulia Davies, Mike Machine Learning Artificial Intelligence Distributed, Parallel, and Cluster Computing Neural and Evolutionary Computing 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. |
| title | Online Continual Learning on Intel Loihi 2 via a Co-designed Spiking Neural Network |
| topic | Machine Learning Artificial Intelligence Distributed, Parallel, and Cluster Computing Neural and Evolutionary Computing |
| url | https://arxiv.org/abs/2511.01553 |