Felix Dangel

BackPACK - a backpropagation package built on top of PyTorch which efficiently computes quantities other than the gradient.

f-dangel

Cockpit: A Practical Debugging Tool for Training Deep Neural Networks

f-dangel

(N=1,2,3)-dimensional unfold (im2col) and fold (col2im) in PyTorch

f-dangel

Hessian backpropagation (HBP): PyTorch extension of backpropagation for block-diagonal curvature matrix approximations

f-dangel

Source code for my PhD thesis: Backpropagation Beyond the Gradient

f-dangel

SINGD: KFAC-like Structured Inverse-Free Natural Gradient Descent (http://arxiv.org/abs/2312.05705)

f-dangel

[TMLR 2022] Curvature access through the generalized Gauss-Newton's low-rank structure: Eigenvalues, eigenvectors, directional derivatives & Newton steps

f-dangel

scipy linear operators for the Hessian, Fisher/GGN, and more in PyTorch

f-dangel