Hossein Mobahi
Scholar

Hossein Mobahi

Google Scholar ID: GSHmKZkAAAAJ
Senior Research Scientist @ Google
Machine Learning
Citations & Impact
All-time
Citations
6,215
 
H-index
23
 
i10-index
29
 
Publications
20
 
Co-authors
41
list available
Resume (English only)
Academic Achievements
  • Paper 'Homotopy Analysis for Tensor PCA' accepted at COLT 2017 (with Anima Anandkumar, Yuan Deng, Rong Ge).
  • Organized the NIPS 2016 workshop on nonconvex optimization.
  • Invited talk 'Off The Convex Optimization' at NIPS 2015 workshop on nonconvex optimization.
  • Paper 'Learning with a Wasserstein Loss' accepted at NIPS 2015.
  • Paper 'The Aperture Problem for Refractive Motion' accepted at CVPR 2015.
  • Research on nonconvex optimization featured in MIT News.
  • Published in top-tier venues including AAAI, CVPR, NIPS, COLT on topics such as Gaussian smoothing, RNN training via diffusion, Wasserstein loss, and refractive motion analysis.
Background
  • Research Scientist in the Machine Perception team at Google Research since May 2016.
  • Broadly interested in Artificial Intelligence, specifically at the intersection of Computer Vision, Machine Learning, and Optimization.
  • Research is often guided by mathematical principles.
  • Previously a Postdoctoral Researcher at MIT's CSAIL, affiliated with the Computer Vision and Sensing, Learning, Inference labs.
  • Collaborated with Profs. Bill Freeman and John Fisher.
  • Admin of the Facebook page on Nonconvex Optimization.