Soroush Nasiriany

I am a second year CS PhD student at UT Austin, advised by Professor Yuke Zhu at the Robot Perception and Learning Lab. I'm working towards developing versatile robot agents that can be deployed in the real world, using tools from machine learning, cognitive science, and computer vision.

Previously, I recieved my undergraduate and master's degrees from UC Berkeley, where I was advised by Professor Sergey Levine. During my time at Berkeley, I helped develop the curriculum for CS 189, the machine learning course.

Recent News

Oct 2021 New preprint on agumenting RL with behavior primitives for robot manipulation
Sept 2021 robomimic accepted as an oral presentation at CoRL 2021!
Aug 2021 We released robomimic, a framework for robot learning from demonstration.
June 2021 Presented DisCo RL at ICRA 2021
Aug 2020 Started my PhD at UT Austin!


Augmenting Reinforcement Learning with Behavior Primitives for Diverse Manipulation Tasks
Soroush Nasiriany, Huihan Liu, Yuke Zhu
In submission
What Matters in Learning from Offline Human Demonstrations for Robot Manipulation
Ajay Mandlekar, Danfei Xu, Josiah Wong, Soroush Nasiriany, Chen Wang, Rohun Kulkarni, Li Fei-Fei, Silvio Savarese, Yuke Zhu, Roberto Martín-Martín
Conference on Robot Learning (CoRL), 2021
Oral Presentation
DisCo RL: Distribution-Conditioned Reinforcement Learning for General-Purpose Policies
Soroush Nasiriany*, Vitchyr H. Pong*, Ashvin Nair*, Alexander Khazatsky, Glen Berseth, Sergey Levine
IEEE International Conference on Robotics and Automation (ICRA), 2021
Planning with Goal-Conditioned Policies
Soroush Nasiriany*, Vitchyr H. Pong*, Steven Lin, Sergey Levine
Advances in Neural Information Processing Systems, 2019

Teaching and Service