Center-Manifold Reduction of Learning at Bifurcations: Interference and Rich Learning in Recurrent Neural Networks

#Center-Manifold #Reduction #Learning #Bifurcations #Interference

arXiv:2605.12763v2 Announce Type: replace-cross Abstract: Rich learning in recurrent neural networks often proceeds through sudden transitions in latent dynamics, but there is little theory predicting how gradient descent behaves during these events. We study the local learning geometry near…