John Duchi
Stanford associate professor of Statistics and Electrical Engineering; first author of the Adagrad paper
- Associate Professor of Statistics and Electrical Engineering, Stanford University12
John Duchi is an associate professor of Statistics and Electrical Engineering at Stanford University, with work spanning statistical learning, optimization, information theory, and computation. Duchi's stated goals include large-scale optimization methods that work robustly across machine learning problems and tools to assess the validity of machine-learned systems. Duchi is first author, with Elad Hazan and Yoram Singer, of the 2011 JMLR paper on adaptive subgradient methods, which PyTorch's documentation cites for its Adagrad optimizer, and before joining the Stanford faculty worked at Google as a software engineer and as a summer research intern and engineer.1234
Work
Adaptive Subgradient Methods for Online Learning and Stochastic Optimization (Adagrad) (external site: jmlr.org)Published research
First author, with Elad Hazan and Yoram Singer, of the 2011 Journal of Machine Learning Research paper presenting subgradient methods that adapt to the geometry of the data seen in earlier iterations; PyTorch's documentation for its Adagrad optimizer refers to this paper for details of the algorithm.34
Other documented roles
Roles the sources document but that are not confirmed as current.
This profile covers public professional work only, from the sources below. It is not an endorsement or a ranking. To correct or remove it, use the correction form.