Jie Tan
I joined the Brain team at Google in 2016, working on deep learning, reinforcement learning and robotics. Before that, I was a Member of Technical Staff at the Computational Imaging group at Lytro, working on computer vision, SLAM, light field technology and image processing. I got my PhD of computer science from Georgia Tech in 2015, under the supervision of Greg Turk and Karen Liu.
My research focused on developing computational tools to understand, simulate and control human and animal motions in a complex environment. I developed fast and stable computer programs to simulate complex dynamic systems, such as fluid, soft body and articulated rigid bodies. I applied optimal control and machine learning techniques to enable computers to automatically learn skills inside a complex physical environment.
I am also interested in transferring the control policies that are learned in simulations to real robots. Policies learned in a simulation usually perform poorly on real robots due to the discrepancies between the simulated and the real system. I am developing tools to understand and model such discrepancies. I augmented the physical simulation using real-world data, which not only increases the simulation accuracy, but also improves the real-world performance of the controllers.
My research focused on developing computational tools to understand, simulate and control human and animal motions in a complex environment. I developed fast and stable computer programs to simulate complex dynamic systems, such as fluid, soft body and articulated rigid bodies. I applied optimal control and machine learning techniques to enable computers to automatically learn skills inside a complex physical environment.
I am also interested in transferring the control policies that are learned in simulations to real robots. Policies learned in a simulation usually perform poorly on real robots due to the discrepancies between the simulated and the real system. I am developing tools to understand and model such discrepancies. I augmented the physical simulation using real-world data, which not only increases the simulation accuracy, but also improves the real-world performance of the controllers.
Research Areas
Authored Publications
Sort By
Google
Robotic table wiping via whole-body trajectory optimizationand reinforcement learning
Benjie Holson
Jeffrey Bingham
Jonathan Weisz
Mario Prats
Peng Xu
Thomas Lew
Xiaohan Zhang
Yao Lu
ICRA (2022)
Learning Model Predictive Controllers with Real-Time Attention for Real-World Navigation
Anthony G. Francis
Dmitry Kalashnikov
Edward Lee
Jake Varley
Leila Takayama
Mikael Persson
Peng Xu
Stephen Tu
Xuesu Xiao
Conference on Robot Learning (2022) (to appear)
Learning Semantic-Aware Locomotion Skills from Human Demonstration
Byron Boots
Xiangyun Meng
Yuxiang Yang
Conference on Robot Learning (CoRL) 2022 (2022) (to appear)
PI-ARS: Accelerating Evolution-Learned Visual Locomotion with Predictive Information Representations
Ofir Nachum
International Conference on Intelligent Robots and Systems (IROS) (2022)
Safe Reinforcement Learning for Legged Locomotion
Jimmy Yang
Peter J. Ramadge
Sehoon Ha
International Conference on Robotics and Automation (2022) (to appear)
Learning to walk on complex terrains with vision
Ale Escontrela
Erwin Johan Coumans
Peng Xu
Sehoon Ha
Conference on Robotic Learning (2021)
Learning Fast Adaptation with Meta Strategy Optimization
Erwin Johan Coumans
Sehoon Ha
Learning Fast Adaptation with Meta Strategy Optimization (2020)
