second-order-rnns-use-weight-tensors

IN premise — entries/2026/06/21/wiki-Recurrent_neural_network-chunk-5.md

Created 2026-06-21T09:55:53+00:00

Second-order RNNs use weight tensors w_ijk (3-index) instead of matrices, enabling direct mapping to finite-state machines in both training and representation.