Drift-Diffusion Matching: Embedding dynamics in latent manifolds of asymmetric neural networks

ArXi:2602.14885v2 Announce Type: replace-cross Recurrent neural networks (RNNs) provide a theoretical framework for understanding computation in biological neural circuits, yet classical results, such as Hopfield's model of associative memory, rely on symmetric connectivity that restricts network dynamics to gradient-like flows. In contrast, biological networks rich time-dependent behaviour facilitated by their asymmetry. Here we