A multiplication-free spike-time learning rule for SNNs achieves 96.5% MNIST and 84.8% Fashion-MNIST accuracy via event-driven FPGA implementation without multiplications or explicit gradients.
Algorithm and hardware design of discrete-time spiking neural networks based on back propagation with binary activations
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.NE 1years
2026 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
A Multiplication-Free Spike-Time Learning Algorithm and its Efficient FPGA Implementation for On-Chip SNN Training
A multiplication-free spike-time learning rule for SNNs achieves 96.5% MNIST and 84.8% Fashion-MNIST accuracy via event-driven FPGA implementation without multiplications or explicit gradients.