Real-time Continual Learning on Intel Loihi 2
Hajizada, Elvin, Rager, Danielle, Shea, Timothy, Campos-Macias, Leobardo, Wild, Andreas, Hüllermeier, Eyke, Sandamirskaya, Yulia, Davies, Mike
–arXiv.org Artificial Intelligence
AI systems on edge devices face a critical challenge in open-world environments: adapting when data distributions shift and novel classes emerge. While offline training dominates current paradigms, online continual learning (OCL)--where models learn incrementally from non-stationary streams without catastrophic forgetting--remains challenging in power-constrained settings. We present a neuromorphic solution called CLP-SNN: a spiking neural network architecture for Continually Learning Prototypes and its implementation on Intel's Loihi 2 chip. Our approach introduces three innovations: (1) event-driven and spatiotemporally sparse local learning, (2) a self-normalizing three-factor learning rule maintaining weight normalization, and (3) integrated neurogenesis and metaplasticity for capacity expansion and forgetting mitigation. On OpenLORIS few-shot learning experiments, CLP-SNN achieves accuracy competitive with replay methods while being rehearsal-free. CLP-SNN delivers transformative efficiency gains: 70\times faster (0.33ms vs 23.2ms), and 5,600\times more energy efficient (0.05mJ vs 281mJ) than the best alternative OCL on edge GPU. This demonstrates that co-designed brain-inspired algorithms and neuromorphic hardware can break traditional accuracy-efficiency trade-offs for future edge AI systems.
arXiv.org Artificial Intelligence
Nov-4-2025
- Country:
- Asia > Myanmar
- Tanintharyi Region > Dawei (0.04)
- Europe
- Germany > Bavaria
- Upper Bavaria > Munich (0.04)
- Switzerland > Zürich
- Zürich (0.04)
- United Kingdom > England
- Cambridgeshire > Cambridge (0.04)
- Germany > Bavaria
- North America > United States
- California > Santa Clara County > Santa Clara (0.04)
- Asia > Myanmar
- Genre:
- Research Report > New Finding (0.46)
- Industry:
- Education (1.00)
- Technology: