analiza aiqbench
This research addresses catastrophic forgetting in Spiking Neural Networks by integrating experience replay with spike-based saliency maps to select informative image patches. The approach significantly reduces storage requirements while maintaining competitive accuracy on standard benchmarks like Split CIFAR10 and 100 under strict memory constraints. It represents a meaningful step toward energy-efficient, lifelong learning systems for neuromorphic hardware.
Saliency-Guided Replay for Continual Learning in Deep Spiking Neural Networks
arXiv:2610.17998v1 Announce Type: cross Abstract: Spiking Neural Networks (SNNs) are promising for energy-efficient neuromorphic computing, but they still suffer from catastrophic forgetting when learning tasks sequentially. This work investigates the integration of Experience Packing and Replay (EPR) as a memory-efficient continual learning strategy for deep SNNs under low-latency constraints. The proposed framework combines Batch Normalization Through Time stabilized VGG9 SNNs with Spike Activation Maps (SAM), which provide forward-only explanations based on spike activity. These explanations are temporally aggregated into spatial attention maps and used to select the most informative image patches for replay, instead of storing full images. Experiments on Split CIFAR10 and Split CIFAR100 show that SAM-guided EPR improves the accuracy-memory trade-off under tight memory budgets, and remains competitive while requiring substantially less storage.
Abstract:Spiking Neural Networks (SNNs) are promising for energy-efficient neuromorphic computing, but they still suffer from catastrophic forgetting when learning tasks sequentially. This work investigates the integration of Experience Packing and Replay (EPR) as a memory-efficient continual learning strategy for deep SNNs under low-latency constraints. The proposed framework combines Batch Normalization Through Time stabilized VGG9 SNNs with Spike Activation Maps (SAM), which provide forward-only explanations based on spike activity. These explanations are temporally aggregated into spatial attention maps and used to select the most informative image patches for replay, instead of storing full images. Experiments on Split CIFAR10 and Split CIFAR100 show that SAM-guided EPR improves the accuracy-memory trade-off under tight memory budgets, and remains competitive while requiring substantially less storage.
Streaming Continual Learning Workshop. In post-proceeding of International Workshops of ECML PKDD 2026
Machine Learning (cs.LG); Artificial Intelligence (cs.AI); Computer Vision and Pattern Recognition (cs.CV); Neural and Evolutionary Computing (cs.NE)
arXiv:2610.17998 [cs.LG]
(or arXiv:2610.17998v1 [cs.LG] for this version)
https://doi.org/10.48550/arXiv.2610.17998 Focus to learn more arXiv-issued DOI via DataCite (pending registration)
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.
Text integral sub licența deschisă a publicației · 241 cuvinte
Subiecte conexe
Local Timescale Gates for Timescale-Robust Continual Spiking Neural Networks
arXiv · 9 oct. 2026
Award-Winning Researcher Trains Robots to Make Educated Guesses
IEEE Spectrum · 12 iun. 2026
Attention-based Experience Replay Framework for Continual Learning of Agnostic Time Series Forecasting Models
arXiv · 24 iul. 2026
Humanity’s Last Invention — Richard Socher of Recursive
Latent Space · 14 sept. 2026