Cathy Wu (Hiring PhD/postdocs) (@wucathy) profile picture

Faculty at @MIT. Learning for control and optimization. Reinforcement Learning. Multi-agent systems. Autonomy. Mobility.

Cambridge, MA
Joined April 2007
Photo by Cathy Wu (Hiring PhD/postdocs) (@wucathy)Photo by Cathy Wu (Hiring PhD/postdocs) (@wucathy)Photo by Cathy Wu (Hiring PhD/postdocs) (@wucathy)Photo by Cathy Wu (Hiring PhD/postdocs) (@wucathy)Photo by Cathy Wu (Hiring PhD/postdocs) (@wucathy)Photo by Cathy Wu (Hiring PhD/postdocs) (@wucathy)Photo by Cathy Wu (Hiring PhD/postdocs) (@wucathy)Photo by Cathy Wu (Hiring PhD/postdocs) (@wucathy)Photo by Cathy Wu (Hiring PhD/postdocs) (@wucathy)Photo by Cathy Wu (Hiring PhD/postdocs) (@wucathy)
This project was 4 years in the making and it's finally out! We found that controlling vehicle speeds to mitigate traffic across a city can cut carbon emissions by 11-22%. To do this, we used deep reinforcement learning to optimize one million eco-driving scenarios. ๐Ÿš—๐Ÿค–๐Ÿง 
1
2
18
1,847
Cathy Wu (Hiring PhD/postdocs) retweeted
๐Ÿš€The era of autonomous multi-agent discovery is arriving! @karpathy ๐ŸชธExcited to share CORAL, our new work on autonomous multi-agent systems for open-ended scientific discovery. ๐Ÿ™…โ€โ™‚๏ธA key limitation of many current โ€œself-evolvingโ€ frameworks is that agents still operate inside tightly constrained loops โ€” they mutate solutions, but they do not truly decide how to explore. In CORAL, we push toward genuine autonomy: Agents decide ๐Ÿ” what to explore ๐Ÿง  what knowledge to store โ™ป๏ธ which ideas to reuse ๐Ÿงช when to test hypotheses ๐Ÿ”ฅOne of the most interesting findings: A single autonomous agent already outperforms fixed evolutionary search, but the biggest gains emerge when multiple agents form a research community. ๐Ÿ’ชOver 50% of breakthroughs in multi-agent runs come from building on other agentsโ€™ discoveries. This suggests that knowledge reuse and collaboration are central to scalable automated discovery. ๐Ÿ…Across 10+ difficult tasks in algorithmic discovery and system optimization, CORAL achieves state-of-the-art performance while improving efficiency by 3โ€“10ร—. ๐Ÿ“„ Paper: arxiv.org/abs/2604.01658v1 ๐Ÿ’ป Code: github.com/Human-Agent-Socieโ€ฆ ๐Ÿ’กAlphaXiv: alphaxiv.org/abs/2604.01658 #agentic #llms #selfevolvingagent #multiagent #autoresearch #alphaevolve
18
112
11
712
55,902
If pure AI or pure search isn't working for you, consider learning-guided optimization! This approach combines the best of both worlds; our most recent work demonstrates this for coordinating agents in warehouse robotics.
1
16
1,077
This wouldn't have been possible without an amazing team. Huge shoutout to Han Zheng (lead author), Yining Ma, Brandon Araki, Jingkai Chen, and Symbotic! #DeepRL #Robotics #WarehouseAutomation #AI
186
Have you noticed how navigation apps include walking & waiting for public transit, but excludes parking & walking for driving? After being late a few times ๐Ÿ˜…, we finally did. We got curious: what if these apps account for parking?
1
1
7
360
Our resulting paper, Probability-Aware Parking Selection, shows that doing so can improve travel time estimates, save drivers time by routing them to parking, and make for fairer comparisons across transportation modes! And this work is featured today on the MIT Home Page!
1
2
216
๐Ÿ“ฃ ๐—ฃ๐—ผ๐˜€๐˜๐—ฑ๐—ผ๐—ฐ ๐—ผ๐—ฝ๐—ฒ๐—ป๐—ถ๐—ป๐—ด๐˜€ ๐—ถ๐—ป ๐—บ๐˜† ๐—ด๐—ฟ๐—ผ๐˜‚๐—ฝ! Iโ€™m looking to hire ๐Ÿญโ€“๐Ÿฎ ๐—ฝ๐—ผ๐˜€๐˜๐—ฑ๐—ผ๐—ฐ๐˜€ to work with us across the groupโ€™s research areas, including: ๐Ÿค– Deep reinforcement learning ๐Ÿšฆ Traffic modeling & control ๐Ÿง  Neural combinatorial optimization
1
1
12
663
Join us in our mission to use ๐—”๐—œ/๐— ๐—Ÿ ๐˜๐—ผ ๐˜๐—ฎ๐—ฐ๐—ธ๐—น๐—ฒ ๐—ต๐—ฎ๐—ฟ๐—ฑ ๐—ผ๐—ฝ๐˜๐—ถ๐—บ๐—ถ๐˜‡๐—ฎ๐˜๐—ถ๐—ผ๐—ป ๐—ฝ๐—ฟ๐—ผ๐—ฏ๐—น๐—ฒ๐—บ๐˜€ and to enable ๐—ฏ๐—ฒ๐˜๐˜๐—ฒ๐—ฟ, ๐—ฒ๐˜ƒ๐—ถ๐—ฑ๐—ฒ๐—ป๐—ฐ๐—ฒ-๐—ฑ๐—ฟ๐—ถ๐˜ƒ๐—ฒ๐—ป ๐—ฑ๐—ฒ๐—ฐ๐—ถ๐˜€๐—ถ๐—ผ๐—ป๐˜€ ๐—ถ๐—ป ๐—ฟ๐—ฒ๐—ฎ๐—น-๐˜„๐—ผ๐—ฟ๐—น๐—ฑ ๐˜€๐—ผ๐—ฐ๐—ถ๐—ผ๐˜๐—ฒ๐—ฐ๐—ต๐—ป๐—ถ๐—ฐ๐—ฎ๐—น ๐˜€๐˜†๐˜€๐˜๐—ฒ๐—บ๐˜€.
1
365
๐Ÿ—“๏ธ Applications are reviewed ๐—ผ๐—ป ๐—ฎ ๐—ฟ๐—ผ๐—น๐—น๐—ถ๐—ป๐—ด ๐—ฏ๐—ฎ๐˜€๐—ถ๐˜€, starting immediately. ๐Ÿ”— Details + how to apply: cathywu.github.io/prospectivโ€ฆ Please share with anyone who might be a great fit!
324
๐Ÿ“ฃ Inviting all in the neural combinatorial optimization community to submit to our upcoming NeurIPS workshop!
789
This work would not have been possible without lead author @vindulajay and collaborators Baptiste Freydt, Ao Qu, Cameron Hickert, Edgar Sanchez, Catherine Tang, Mark Taylor, Blaine Leonard, and financial support of UDOT, @NSF, and Amazon AWS, and support from MIT SuperCloud.
247
Are you a transportation practitioner or researcher who has read "Confessions of a Recovering Engineer" by Charles Marohn? And attending #TRBAM in a couple weeks? I found this scathing critique of traffic engineering to be thought provoking, and would love to discuss it.
1
2
3
1,433
Iโ€™d love to have a constructive conversation about the future of traffic engineering, needs of the profession, and educating the next generation of engineers. If there is enough interest, Iโ€™m happy to do a bit of a book club at TRB. Feel free to share the invite with friends.
2
2
648
Cathy Wu (Hiring PhD/postdocs) retweeted
The Climate Change AI workshop at NeurIPS tomorrow will be livestreamed for free, even if you weren't able to register for NeurIPS (which was oversubscribed this year)!
Another highlight of our #NeurIPS2024 workshop on Sunday will be a panel discussion on small vs big models in ML for climate impact, featuring Chris Bretherton, Anamika Dubey & @BDilkina. Stream the whole workshop live & for free via our website: climatechange.ai/events/neurโ€ฆ
12
44
26,408
Want to learn more about neural combinatorial optimization (NCO)? Informal meetup in an hour at #NeurIPS2024, followed by lunch! 11am, West exhibition hall, near Door A. #ORMS Details: calendar.app.google/NEd7HMqmโ€ฆ
1
5
693