Gaspard Goupy

Gaspard Goupy

PhD student working on spiking neural networks

Paired competing neurons improving STDP supervised local learning in spiking neural networks

Published in Frontiers in Neuroscience, 2024

Paper - Code - Slides

In this paper, we introduce:

  1. Stabilized Supervised STDP (S2-STDP), a learning rule for training a spiking classification layer employing first-spike coding and single-spike integrate-and-fire neurons;

  2. Paired Competing Neurons (PCN), an architecture based on intra-class winner-takes-all competition that promotes neuron specialization toward target or non-target samples.

Figure of the PCN architecture

Recommended citation: Gaspard Goupy, Pierre Tirilly, and Ioan Marius Bilasco. Paired Competing Neurons Improving STDP Supervised Local Learning in Spiking Neural Networks. Frontiers in Neuroscience, 18, 2024.