TorchLean API

NN.MLTheory.Proofs.Hopfield.Dynamics

Hopfield global dynamics: energy along trajectories #

This file lifts the single-step energy lemma to whole trajectories: for any (possibly adversarial) asynchronous update schedule useq, the energy sequence is non-increasing.

Stronger theorems from the paper (convergence for fair schedules; cyclic convergence bounds) can be built on top of these monotonicity facts.

Energy is nonincreasing along any update sequence, one step at a time.

Hence energy never exceeds its starting value, for any schedule of units.

The schedule is an arbitrary NatFin n, so this covers random and cyclic orders alike.