Aviv Tamar

Welcome to my homepage!

I’m an assistant professor at Technion – Israel Institute for Technology, in the Electrical and Computer Engineering (ECE) department.

My research focuses on AI and machine learning, with an emphasis on robotics applications. My long term goal is to bring robots into human-centered domains such as homes and hospitals. Towards this goal, some fundamental questions need to be solved, such as how can machines learn models of their environments that are useful for performing tasks, and how to learn behavior from interaction in an interpretable and safe manner. Most of my work falls under the framework of reinforcement learning, and its connections to representation learning, planning, and risk-averse optimization. See my publications page for more details!

Previously, I was a postdoc in the Berkeley AI Research Lab (BAIR) at UC Berkeley, with Prof. Pieter Abbeel. I completed my PhD. at the Technion, supervised by Prof. Shie Mannor, and my MSc also at Technion, under the supervision of Prof. Ron Meir.

News

1. Mar 2022

Our work Validate on Sim, Detect on Real--Model Selection for Domain Randomization got accepted to ICRA 2022

1. Jan 2022

Won ERC Starting grant for project BAYES-RL!

18. Dec 2021

We're looking for a postdoc to join our team! Contact me if interested.

16. Dec 2021

Regularization Guarantees Generalization in Bayesian Reinforcement Learning through Algorithmic Stability accepted to AAAAI 2021!

16. Dec 2021

New preprint: Validate on Sim, Detect on Real--Model Selection for Domain Randomization

11. Oct 2021

New preprint: Regularization Guarantees Generalization in Bayesian Reinforcement Learning through Algorithmic Stability

11. Oct 2021

Offline meta learning of exploration got accepted to NeurIPS 2021!

1. Apr 2021

2 papers (contrastive domain randomization and efficient self-supervised data collection) got accepted to ICRA 2021, and Soft-Intro VAE accepted as an oral presentation at CVPR 2021!

3. Sep 2020

New preprint: offline meta reinforcement learning

... see all News