Tag
This paper extends Equilibrium Propagation to skew-gradient systems and demonstrates an equivalence between deep Energy-Based Models and Hamiltonian neural networks, focusing on diffusively coupled Fitzhugh-Nagumo neurons. It derives a layer-wise Hamiltonian recurrence relation for inference in such networks.
This paper investigates integrating dendritic neural networks with equilibrium propagation, showing that this biologically plausible approach improves performance on challenging datasets compared to standard equilibrium propagation.