I am a fifth-year Ph.D. candidate at MIT EECS where I am fortunate to be co-advised by Daniel Freund and Thodoris Lykouris at MIT Sloan.

My research optimizes human decision-making for safer societies. Combining tools from machine learning and queueing theory, I design data-driven algorithms and provide practical insights for content moderation and the adjudication of legal proceedings. Three of my papers have been featured in INFORMS paper competitions, including the 2024 Public Sector Operations Research Best Paper Competition (2nd Place), the 2024 Service Science Best DEIJ Paper Award (1st Place), the 2024 Junior Faculty Interest Group Paper Competition (finalist), and the 2022 Applied Probability Society Best Student Paper Competition (finalist).

In the Summer of 2024, I was a research scientist intern with Central Applied Science at Meta, developing queue prioritization and measurement algorithms for content moderation. I obtained a B.E. in Computer Science in 2021 from Yao Class, Tsinghua University.

I am on the 2025-2026 academic job market!

Email: wweng (at) mit.edu

[Google scholar] [CV (last updated: Oct 25)]

Scheduled and Recent Talks

Learning to Defer in Congested Systems: The AI-Human Interplay, INFORMS Annual Meeting, October 2025
- October 29, 8:00AM - 8:15AM at Building A, Level 4, A405 (session: Learning and Control of Stochastic Systems)
Scheduling with Uncertain Holding Costs and its Application to Content Moderation, Cornell Young Researchers Workshop, October 2025
Regulating Wait-Driven Requests in Queues, ACM EC, INFORMS APS, July 2025
Learning to Defer in Congested Systems: The AI-Human Interplay, INFORMS APS, July 2025

Research

Working Papers

Scheduling with Uncertain Holding Costs and its Application to Content Moderation

with Caner Gocmen, Thodoris Lykouris, Deeksha Sinha, Major Revision at Management Science.

Regulating Wait-Driven Requests in Queues

with Daniel Freund, David Hausman, under review at Management Science.

- - A version of this paper was accepted for presentation at EC 2025, MIW 2025
The Dedicated Docket in U.S. Immigration Courts: An Analysis of Fairness and Efficiency Properties

with Daniel Freund, under second-round review at Manufacturing & Service Operations Management after Major Revision.

A version of this paper was accepted for presentation at EC 2024, MSOM 2024, EAAMO 2024
Second place for the 2024 Public Sector Operations Research Best Paper Competition
First place for the 2024 Service Science Best DEIJ Paper Award

Learning to Defer in Congested Systems: The AI-Human Interplay

with Thodoris Lykouris, under second-round review at Operations Research after Major Revision.

A version of this paper was accepted for presentation at MSOM 2024, MIW 2024
Finalist for the 2024 Junior Faculty Interest Group Paper Competition

The Transient Cost of Learning in Queueing Systems

with Daniel Freund and Thodoris Lykouris, under second-round review at Operations Research after Reject & Resubmit.

- - An earlier version of this paper was accepted at NeurIPS 2023, titled ``Quantifying the Cost of Learning in Queueing Systems''
Group Fairness in Dynamic Refugee Assignment

with Daniel Freund, Thodoris Lykouris, Elisabeth Paulson, and Bradley Sturt, Major Revision at Operations Research.

A version of this paper was accepted for presentation at EC 2023, FORC 2023, MIW 2023

Journal Papers

Efficient Decentralized Multi-Agent Learning in Asymmetric Bipartite Queueing Systems

with Daniel Freund, Thodoris Lykouris, Operations Research. Vol. 72, No. 3 (2024): 1049-1070.

An earlier version of this paper was accepted at COLT 2022
Finalist for the 2022 Applied Probability Society Best Student Paper Competition

An Algorithm for Improved Delay-Scaling in Input Queued Switches

with R. Srikant, Queueing Systems, Vol. 100 (2022): 135-166.

Refereed conference publications not listed above

Labor-Right Protecting Dispatch of Meal Delivery Platforms

with Yang Yu, IEEE Conference on Decision and Control, 2021. [arXiv Preprint]

Optimal Load Balancing with Locality Constraints

with Xingyu Zhou, R. Srikant, ACM SIGMETRICS, 2021. [arXiv Preprint]

Achieving Zero Asymptotic Queueing Delay for Parallel Jobs

with Weina Wang, ACM SIGMETRICS, 2021. [arXiv Preprint]

The Mean-Squared Error of Double Q-Learning

with Harsh Gupta, Niao He, Lei Ying, R. Srikant, NeurIPS, 2020. [arXiv Preprint]

Teaching

Teaching Assistant:
- Reinforcement Learning: Foundations and Methods, Fall 2024 (TA evaluation 7.0/7.0)
- The Theory of Operations Management, Spring 2024 (TA evaluation 6.9/7.0)
- Introduction to Operations Management, Spring 2023 (TA evaluation 6.2/7.0)
Guest lecturer:
- Law, Ethics, and Data Privacy, hosted by Jordan Levine, MIT Sloan, January 2025.
  - 25min lecture based on my work Group Fairness in Dynamic Refugee Assignment.
- Queueing Models for Service Operations Management, hosted by Prof. Amy Ward, Chicago Booth, October 2024.
  - 90min lecture based on my work The Dedicated Docket in U.S. Immigration Courts: An Analysis of Fairness and Efficiency Properties.