3 citations · 11 across the 23 of their papers we have counts for
Showing 2021Show all
2 papers · 1 filter
cs.LG2021★ 1 cited
ViViT: Curvature access through the generalized Gauss-Newton's low-rank structure
Felix Dangel, Lukas Tatzel, Philipp Hennig
Curvature in form of the Hessian or its generalized Gauss-Newton (GGN) approximation is valuable for algorithms that rely on a local model for the loss to train, compress, or expla…
cs.LG2021★ 3 cited
Cockpit: A Practical Debugging Tool for the Training of Deep Neural Networks
Frank Schneider, Felix Dangel, Philipp Hennig
When engineers train deep learning models, they are very much 'flying blind'. Commonly used methods for real-time training diagnostics, such as monitoring the train/test loss, are…