Kevin Berlemont

Glassy phase in dynamically-balanced neural networks

The paper introduces a mean-field theory for balanced neuronal networks, identifying a glassy phase with non-exponential relaxation and broken ergodicity beyond a critical synaptic gain.

Confidence-controlled Hebbian learning efficiently extracts category membership from stimuli encoded in view of a categorization task

The paper introduces a confidence-controlled Hebbian learning rule that efficiently extracts category membership from optimized stimulus encoding layers, outperforming reward-modulated Hebbian learning.

Nonlinear neural network dynamics accounts for human confidence in a sequence of perceptual decisions

This study shows that an attractor neural network model can reproduce human confidence judgments in perceptual decisions, linking confidence to the nonlinear dynamics of the decision-making process.