David P. Helmbold
Identifiers
- name variant David P. Helmbold 0.60 · backfill
Papers (5)
- Gradient descent with identity initialization efficiently learns positive definite linear transformations by deep residual networks cs.LG · 2018 · author #2
- Online Learning of Combinatorial Objects via Extended Formulation cs.LG · 2016 · author #2
- Surprising properties of dropout in deep networks cs.LG · 2016 · author #1
- On the Inductive Bias of Dropout cs.LG · 2014 · author #1
- On the Necessity of Irrelevant Variables cs.LG · 2012 · author #1
Mentions
Frequent Coauthors
- Philip M. Long 4 shared papers
- Holakou Rahmanian 1 shared papers
- Peter L. Bartlett 1 shared papers
- S.V.N. Vishwanathan 1 shared papers