Thank you for your interest in our lab!

How to apply
  • Prospective PhD students: I take new PhD students each year. I do not admit PhD students directly; if you are interested in a PhD position, apply to the CEE, SES (IDSS), ORC, and/or EECS PhD programs and list me in your application, so that your application is directed to me. I am also open to advising students from other programs with the right fit. I ask that you do not contact me directly with regard to PhD admissions until after you are admitted, as I will not be able to reply to emails from individual applicants.
  • Prospective postdocs: If you are interested in a postdoc position, please read this form.
  • MIT undergraduate or Masters students: If you are interested in research positions, please read this form.
  • Potential interns and visitors: If you are not a MIT student and interested in visiting research positions, please read this form.

If you cannot view the form (because you don’t have a Google account), you may request an application by email with the subject heading “[Application Request]”. Please specify which application form you need.

If your case is not covered by the above, then consider cold emailing me!

I especially encourage those who are underrepresented in STEM to apply. I greatly value diversity and inclusion in STEM through teaching, mentorship, and outreach. We are a highly collaborative and welcoming lab with diverse backgrounds that come together to make the whole stronger than the sum of its parts. I strongly believe that diversity in perspective is necessary to solve the world’s most challenging problems.

Topic areas

We are recruiting in the following topic areas. For more information, see the listed example papers. We are open to applicants at all levels. If any of these topics excite you, please reference them in your application.

  • Neural combinatorial optimization, i.e. machine learning for combinatorial optimization, including end-to-end learning, foundation models, learning-guided optimization, and applications
    • Learning to Configure Separators in Branch-and-Cut (NeurIPS 2023)
    • Learning to Delegate for Large-scale Vehicle Routing (NeurIPS 2021 Spotlight)
  • Generalization in reinforcement learning, including contextual reinforcement learning, multi-task learning, transfer learning, and applications
    • Model-Based Transfer Learning for Contextual Reinforcement Learning (NeurIPS 2024)
    • The Impact of Task Underspecification in Evaluating Deep Reinforcement Learning (NeurIPS 2022)
  • (New) Naturalistic traffic simulation, including long-term traffic simulation, microsimulation calibration, driver interaction modeling
  • Sustainable mobility, including optimization for energy-efficient driving (eco-driving), congestion mitigation, and transit operations
    • Cooperative Advisory Residual Policies for Congestion Mitigation (ACM JATS 2024)
    • Mitigating Metropolitan Carbon Emissions with Semi-autonomous Vehicles using Deep Reinforcement Learning (pre-print available)
  • Multi-agent coordination, particularly for 10-1000 agents, including multi-agent path finding (MAPF), multi-robot motion planning (MRMP), task assignment (TA), mobile robots, human interaction, and applications (especially in mixed autonomy traffic and autonomous warehousing)
    • Neural Neighborhood Search for Multi-agent Path Finding (ICLR 2024)
    • Multi-agent Path Finding for Cooperative Autonomous Driving (ICRA 2024)
  • Road safety, including proactively measuring and co-designing for system safety
    • Revisiting the Correlation between Simulated and Field-Observed Conflicts Using Large-Scale Traffic Reconstruction (AAP 2024, to appear)
    • Cooperation for Scalable Supervision of Autonomy in Mixed Traffic (T-RO 2023)
  • (New) Offline reinforcement learning, including hybrid RL and applications
  • (New) Reproducible research, including quantifying (non-)reproducibility in engineering