Coline Devin

I am a senior research scientist at DeepMind. I completed my PhD in 2020 at UC Berkeley, advised by Professors Sergey Levine, Pieter Abbeel and Trevor Darrell, and supported by the NSF Graduate Research Fellowship. I have previously interned at OSARO, Google Brain, and DeepMind. I completed my B.S. in Computer Science at Harvey Mudd College.

Google Scholar  /  GitHub  /  Dissertation Talk


5th Inter-experiment Machine Learning Workshop and EP-IT Data science seminar: Beyond Pick and Place - Tackling Robotic Stacking of Diverse Shapes. 2022

Research and Applied AI Summit: Modularity for Robot Learning. 2019


My research is robot learning, with an emphasis on approaches that can show compositionality and generality in tasks, objects, and environments.

How to Spend Your Robot Time: Bridging Kickstarting and Offline Reinforcement Learning for Vision-based Robotic Manipulation
Alex X Lee*, Coline Devin*, Jost Tobias Springenberg*, Yuxiang Zhou, Thomas Lampe, Abbas Abdolmaleki, Konstantinos Bousmalis
in submission

Beyond pick-and-place: Tackling robotic stacking of diverse shapes
Alex X Lee*, Coline Devin*, Yuxiang Zhou*, Thomas Lampe*, Konstantinos Bousmalis*, Jost Tobias Springenberg*, Arunkumar Byravan, Abbas Abdolmaleki, Nimrod Gileadi, David Khosid, Claudio Fantacci, Jose Enrique Chen, Akhil Raju, Rae Jeong, Michael Neunert, Antoine Laurens, Stefano Saliceti, Federico Casarini, Martin Riedmiller, Francesco Nori
CoRL 2021
arxiv / blog / video

Fully Autonomous Real-World Reinforcement Learning with Applications to Mobile Manipulation
Charles Sun, J╚ędrzej Orbik, Coline Manon Devin, Brian H Yang, Abhishek Gupta, Glen Berseth, Sergey Levine
CoRL 2021

SMiRL: Surprise Minimizing RL in Dynamic Environments
Glen Berseth, Daniel Geng, Coline Devin, Chelsea Finn, Dinesh Jayaraman, Sergey Levine
ICLR 2021 (Oral)
arxiv / blog

Learning To Reach Goals Without Reinforcement Learning
Dibya Ghosh, Abhishek Gupta, Justin Fu, Ashwin Reddy, Coline Devin, Benjamin Eysenbach, Sergey Levine

Plan Arithmetic: Compositional Plan Vectors for Multi-Task Control
Coline Devin, Daniel Geng, Pieter Abbeel, Trevor Darrell, Sergey Levine
NeurIPS 2019
arxiv / webpage / short video / code

Monocular Plan View Networks for Autonomous Driving
Dequan Wang, Coline Devin, Qi-Zhi Cai, Philipp Krahënbühl, Trevor Darrell
IROS 2019

Deep Object Centric Policies for Autonomous Driving
Dequan Wang, Coline Devin, Qi-Zhi Cai, Fisher Yu, Trevor Darrell
ICRA 2019

Grasp2Vec: Learning Object Representations from Self-Supervised Grasping
Eric Jang*, Coline Devin*, Vincent Vanhoucke, Sergey Levine
* Denotes equal contribution
CoRL 2018
arxiv / webpage / code

Deep Object-Centric Representations for Generalizable Robot Learning
Coline Devin, Pieter Abbeel, Trevor Darrell, Sergey Levine
ICRA 2018
arxiv / webpage / code

Learning Invariant Feature Spaces to Transfer Skills with Reinforcement Learning
Abhishek Gupta*, Coline Devin*, YuXuan Liu, Pieter Abbeel, Sergey Levine
* Denotes equal contribution
ICLR, 2017
webpage / video

Learning Modular Neural Network Policies for Multi-Task and Multi-Robot Transfer
Coline Devin*, Abhishek Gupta*, Trevor Darrell, Pieter Abbeel, Sergey Levine
* Denotes equal contribution
ICRA, 2017
arxiv / webpage / video

Adapting deep visuomotor representations with weak pairwise constraints
Eric Tzeng*, Coline Devin*, Judy Hoffman, Chelsea Finn, Pieter Abbeel, Sergey Levine, Kate Saenko, Trevor Darrell
WAFR, 2016

Website template from Jon Barron.
Last updated June 2022.