← Search

Bhargav Kanuparthi

2 accepted papers

2020

Untangling tradeoffs between recurrence and self-attention in artificial neural networks

NeurIPS 2020poster

Attention and self-attention mechanisms, are now central to state-of-the-art deep learning on sequential tasks. However, most recent progress hinges on heuristic approaches with limited understanding of attention's role in model optimization and computation, and rely on considerable memory and compu…

2019

h-detach: Modifying the LSTM Gradient Towards Better Optimization

ICLR 2019poster

Recurrent neural networks are known for their notorious exploding and vanishing gradient problem (EVGP). This problem becomes more evident in tasks where the information needed to correctly solve them exist over long time scales, because EVGP prevents important gradient components from being back-pr…