Alexei A. Efros
Howard Friesen Professor in UC Berkeley EECS and member of BAIR; co-author of pix2pix and CycleGAN
- Howard Friesen Professor, EECS Department; member of BAIR, University of California, Berkeley1
Alexei A. Efros (Alyosha) is the Howard Friesen Professor in the EECS Department at UC Berkeley and part of the Berkeley Artificial Intelligence Research Lab (BAIR), after a decade on the faculty of the Robotics Institute at Carnegie Mellon University. Efros's research uses large amounts of unlabeled visual data to understand, model, and recreate the visual world, mainly in data-driven computer vision and its links to computer graphics and computational photography, with interests that include self-supervised learning and world models. Efros is a co-author of the image-to-image translation papers known as pix2pix (2016) and CycleGAN (2017), whose public code repositories list Efros among the authors.12345
Work
pix2pix (image-to-image translation with conditional adversarial networks) (external site: arxiv.org)Published research
Co-author, with Phillip Isola, Jun-Yan Zhu, and Tinghui Zhou, of "Image-to-Image Translation with Conditional Adversarial Networks" (2016); the pix2pix code repository lists Efros among its authors.23
CycleGAN (external site: arxiv.org)Published research
Co-author, with Jun-Yan Zhu, Taesung Park, and Phillip Isola, of "Unpaired Image-to-Image Translation using Cycle-Consistent Adversarial Networks" (2017); the PyTorch CycleGAN and pix2pix repository lists Efros among the authors of both papers.45
Other documented roles
Roles the sources document but that are not confirmed as current.
- Faculty, Robotics Institute (a decade, before Berkeley), Carnegie Mellon University1
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.