Alexander Sax

Alexander (Sasha) Sax

PhD Student, Computer Science

UC Berkeley

About me:

My research interests are in developing simple models of visual perception. I ultimately want to understand how these systems are shaped by their ecological function. My current research angle is to try to model the relationships between different visual tasks in order to understand some of the general principles. My work so far has been modeling relationships between different computer vision tasks, and then linking those vision tasks to more ecologically realistic tasks in robotics.

I’m currently at UC Berkeley, where I am a PhD student advised by Jitendra Malik. I’m also advised by Amir Zamir (@ EPFL). Before coming to Berkeley, I was at Stanford for a M.Sc. (in Computer Science advised by Silvio Savarese and Amir Zamir and B.Sc. (in Math), also from Stanford.

Selected Honors and Awards

Education

  • PhD in EECS, 2022

    UC Berkeley

  • MSc in CS, 2018

    Stanford University

  • BSc in Math, 2018

    Stanford University

Selected Publications

Updated version on Google scholar

Complete list on Google scholar.
Robustness via Cross-Domain Ensembles. In ICCV, 2021.  [Oral]

PDF Code Project Project Site

Robust Policies via Mid-Level Visual Representations. In CoRL, 2020.  

PDF Code Project Video Project Site

Robust Learning Through Cross-Task Consistency. In CVPR, 2020.  [Best Paper Nominee] [Oral]

PDF Code Project Project Site

Side-Tuning: A Baseline for Network Adaptation via Additive Side Networks. In CVPR, 2020.  [Oral]

PDF Code Project Project Site

Mid-Level Visual Priors Improve Generalization and Sample Efficiency for Learning Visuomotor Policies. In CoRL, in BayLearn, 2019.  [Oral]

PDF Code Project Video Project Site

Education

Internships and Research Positions