I am interested in developing algorithms and systems to enable reliable decision-making in urban and societal systems. Directions of research include sample-efficient reinforcement learning, distribution shift, bridging machine learning and automation science, and automation science in the context of automated vehicles. My research is motivated by the challenge of understanding and shaping the impact of autonomy on society. Ultimately, this research will inform complex decision-making, from automated vehicles and transportation systems to urban planning and public policy.

I am a postdoc with Microsoft Research AI. I will join the MIT faculty in Fall 2019, where I will be part of the Department of Civil and Environmental Engineering (CEE), the Institute for Data, Systems, & Society (IDSS), and the Laboratory for Information & Decision Systems (LIDS).

I recently completed my PhD in EECS at UC Berkeley, working at the intersection of machine learning, optimization, and mobility. My PhD research focused on mixed autonomy systems in mobility, which studies the complex integration of automation such as self-driving cars into existing urban systems. I was advised by Alexandre Bayen, and was part of the Berkeley AI Research Lab (BAIR), California Partners for Advanced Transportation Technology (PATH), the Berkeley Real-time Intelligent Secure Explainable Systems Lab (RISELab), and Berkeley DeepDrive (BDD). Before graduate school, I received a BS and MEng in EECS from MIT, where I worked with Daniela Rus, Seth Teller, and Jim Glass. I have also spent time at OpenAI, Microsoft Research, the Google X Self-Driving Car Team (now Waymo), Dropbox, Facebook, and several startups.

Recent news

  • March 2019: I am honored to be inducted into the Microsoft Location Summit Hall of Fame, for receiving the First Place Award for my recent talk on “Integrating Autonomy into Urban Systems.”
  • December 2018: I am honored to receive the 2018 Milton Pikarsky Memorial Award from the Council of University Transportation Centers (CUTC) for my PhD dissertation. See the corresponding Berkeley ITS article here.
  • November 2018: I joined the Reinforcement Learning Group at Microsoft Research AI in Redmond as a postdoctoral researcher.
  • November 2018: The Flow team and I gave a full-day Tutorial and a half-day Workshop on Reinforcement Learning and Transportation at the IEEE Intelligent Transportation Systems Conference (ITSC) in Maui, Hawaii. All tutorial materials are available here.
  • September 2018: I gave an invited talk at the O’Reilly Artificial Intelligence Conference on “Reinforcement Learning for Mixed Autonomy Traffic” about the Flow project.

Selected publications

For full list, see publications.

Learning and Optimization for Mixed Autonomy Systems – A Mobility Context
Cathy Wu
Thesis. PhD, Electrical Engineering and Computer Sciences, UC Berkeley, 2018.

Flow: Architecture and Benchmarking for Reinforcement Learning in Traffic Control
Cathy Wu, Aboudy Kreidieh, Kanaad Parvate, Eugene Vinitsky, Alexandre Bayen
IEEE Transactions on Robotics (T-RO). In review.
arXiv / videos / github / project page

Lagrangian Control through Deep-RL: Applications to Bottleneck Decongestion
Eugene Vinitsky, Kanaad Parvate, Abdul Rahman Kreidieh, Cathy Wu, Alexandre Bayen
IEEE Intelligent Transportation Systems Conference (ITSC), 2018.

Variance Reduction for Policy Gradient Using Action-Dependent Factorized Baselines
Cathy Wu, Aravind Rajeswaran, Yan Duan, Vikash Kumar, Alexandre M Bayen, Sham Kakade, Igor Mordatch, Pieter Abbeel
International Conference on Learning Representations (ICLR), 2018. Oral (2%).
Deep Reinforcement Learning Symposium (NIPS), 2017. Contributed talk.
arXiv / OpenReview

Stabilizing traffic with autonomous vehicles
Cathy Wu, Alexandre Bayen, Ankur Mehta
International Conference on Robotics and Automation (ICRA), 2018.
proceedings / talk video

Emergent behaviors in mixed-autonomy traffic
Cathy Wu, Aboudy Kreidieh, Eugene Vinitsky, Alexandre Bayen
Conference on Robot Learning (CoRL), 2017.
proceedings / pdf

Clustering for Set Partitioning with a Case Study in Ridesharing
Cathy Wu, Ece Kamar, Eric Horvitz
IEEE Intelligent Transportation Systems Conference (ITSC), 2016. Best paper award.
proceedings / pdf

Cellpath: fusion of cellular and traffic sensor data for route flow estimation via convex optimization
Cathy Wu, Jerome Thai, Steve Yadlowsky, Alexei Pozdnoukhov, Alexandre Bayen
Transportation Research: Part C, 2015.
International Symposium on Transportation and Traffic Theory (ISTTT), 2015. Oral (14%).
journal / pdf / github (system) / github (algorithm)


EE290O: Deep multi-agent reinforcement learning with applications to autonomous traffic
Co-instructor, UC Berkeley, Fall 2018.

CS189: Introduction to Machine Learning
Graduate Student Instructor (Content Development), UC Berkeley, Fall 2017.

CS170: Efficient Algorithms and Intractable Problems
Graduate Student Instructor (Head TA), UC Berkeley, Spring 2016.

6.004: Computation Structures
Teaching Assistant, MIT, Spring 2012.

Selected other writing

It’s Time to Do Something: Mitigating the Negative Impacts of Computing Through a Change to the Peer Review Process.
Brent Hecht, Lauren Wilcox, Jeffrey P. Bigham, Johannes Schöning, Ehsan Hoque, Jason Ernst, Yonatan Bisk, Luigi De Russis, Lana Yarosh, Bushra Anjum, Danish Contractor, Cathy Wu
ACM Future of Computing Academy Blog, 2018.
full article

Traffic Jammin’: Making automated transportation a reality
Cathy Wu
Berkeley Science Review, 2016.
full article

Automating us: The entanglement of people and machines
Daniel Aranki*, Roel Dobbe*, Jaime F. Fisac*, and Cathy Wu*
Berkeley Science Review, 2015.
full article

How to Lead a Technical Reading Group
Cathy Wu, Oct 2012.