Browse Teach Curate

Unrolling a Recurrence into a Deep Computational Graph

reframing a loop as a depth-T network is the move that makes both the gradient pathology and the parallelism barrier visible

jeremy Video

Recurrent Neural Networks in Machine Learning for Predicting Sequential Stock Prices

Explains RNNs for variable-length sequences: feedback loops carrying a hidden state, unrolling the recurrence, shared weights across time steps, and the vanishing/exploding gradient problem that moti…