multi-plasticity-snn-training

What problem does it solve?

Coordinating multiple synaptic plasticity mechanisms to train spiking neural networks efficiently, enabling better adaptation and learning in neuromorphic contexts.

Core Features & Use Cases

• Multi-plasticity coordination (STDP, reward modulation, Hebbian learning) for end-to-end SNN training.
• Adaptive mechanism weighting to allocate learning updates per task and data characteristics.
• PyTorch-based implementation suitable for online learning and low-power AI experiments.
• Use cases include neuromorphic computing, temporal sequence learning, and online adaptation in constrained environments.

Quick Start

Train a multi-plasticity SNN on a temporal dataset using the provided PyTorch module.