Felix Dangel
Scholar

Felix Dangel

Google Scholar ID: 9hlJ9W0AAAAJ
Postdoc at the Vector Institute, Toronto
Second-order optimizationautomatic differentiationdeep neural networkstensor networks
Homepage↗Google Scholar↗
Citations & Impact
All-time
Citations
396
 
H-index
11
 
i10-index
11
 
Publications
20
 
Co-authors
28
list available
Contact
Emailfelix.dangel@vectorinstitute.aiGitHubOpen ↗
Publications
12 items
Efficient Bilevel Optimization with KFAC-Based Hypergradients
2026
Cited
0
Dataless Weight Disentanglement in Task Arithmetic via Kronecker-Factored Approximate Curvature
2026
Cited
0
Sketching Low-Rank Plus Diagonal Matrices
2025
Cited
0
Understanding and Improving the Shampoo Optimizer via Kullback-Leibler Minimization
2025
Cited
0
Fishers for Free? Approximating the Fisher Information Matrix by Recycling the Squared Gradient Accumulator
2025
Cited
0
Kronecker-factored Approximate Curvature (KFAC) From Scratch
2025
Cited
0
Collapsing Taylor Mode Automatic Differentiation
2025
Cited
0
Improving Energy Natural Gradient Descent through Woodbury, Momentum, and Randomization
2025
Cited
0
Resume (English only)
Co-authors
28 total
Philipp Hennig
Philipp Hennig
University of Tübingen
Runa Eschenhagen
Runa Eschenhagen
PhD student, University of Cambridge
Agustinus Kristiadi
Agustinus Kristiadi
Assistant Professor, Western University
Co-author 4
Co-author 4
Co-author 5
Co-author 5
Richard E Turner
Richard E Turner
Professor, University of Cambridge
Co-author 7
Co-author 7
Frederik Kunstner
Frederik Kunstner
INRIA