About myself

I study hidden structure in networks and how to make better decisions in matching markets.

I am an assistant professor of Operations, Information and Decisions at the Wharton School of Business. I was a Postdoctoral Scholar under Prof. Itai Ashlagi and Prof. Amin Saberi, at Management Science and Engineering, Stanford University. I received my PhD in Decision Sciences from the Fuqua School of Business at Duke University, under Prof. Jiaming Xu and Prof. Yehua Wei.

My research interests focus on data analysis, algorithm design, and performance evaluation in large-scale networks and stochastic systems. My works draw inspiration from real-world business, engineering, and natural sciences problems that can be modeled into large and complex networks. I have explored a range of topics from the fundamental limits and efficient algorithms on graph matching to online platform policy design with bounded regret and data confidentiality protection.

News

• Can you design a better hiring rule in the age of AI? Play our hiring game to explore how screening rules shape applicant behavior and hiring outcomes, based on our paper Screening AI-equipped applicants.

• I am serving as a council member of the INFORMS Applied Probability Society (2026–2028) and as a cluster co-chair for the INFORMS MSOM Service Management Special Interest Group (SIG) (2026–2027).

• I am chairing the SNAPP (Stochastic Networks, Applied Probability, and Performance) online seminar series, 2026–2027.

• My co-author Taha Ameen is on the 2026-2027 job market.

• Our paper A uniformity principle for spatial matching is selected for Outstanding Student Poster Award at Stochastic Networks Conference (2026) and finalist for the 2026 INFORMS MSOM Student Paper Competition.

• Our paper Constant regret primal-dual policy for multi-way dynamic matching is accepted at Management Science, 2026.

• Our paper Online metric matching: beyond the worst case is accepted at Operations Research, 2026.

• Our paper From signaling to interviews in random matching markets is accepted at STOC, 2025.

• Our paper Random graph matching at Otter's threshold via counting chandeliers is accepted at Operations Research, 2025.

• Our paper Stochastic online metric matching: adversarial is no harder than stochastic is accepted at WINE, 2024.

• I received Thomas M. Cover Dessertation Award from IEEE Information Theory Society, 2024.

• I will give a tutorial on Matching in Networks: Fundamental Limits and Efficient Algorithms with Prof. Jiaming Xu at ACM SIGMETRICS, 2024.

• My dissertation Matching in Networks: Fundamental Limits and Efficient Algorithms received Best Dissertation Award from Fuqua.

• Our paper Constant regret primal-dual policy for multi-way dynamic matching is accepted to SIGMETRICS 2023.

• Our paper Random graph matching at Otter's threshold via counting chandeliers is accepted to STOC 2023.

• Our paper Testing network correlation efficiently via counting trees is accepted to Annals of Statistics.

• Our paper Testing network correlation efficiently via counting trees is selected as a finalist for "George Nicholson Student Paper Competition", INFORMS 2022.

• Our paper Settling the sharp reconstruction thresholds of random graph matching is accepted to IEEE Transactions on Information Theory.

• I am teaching EGRMGMT-580 Decision Modeling (Graduate course) at Pratt school of engineering, Spring 2022.

• Our paper Testing correlation of unlabeled random graphs is accepted to Annals of Applied Probability.

• I visited Simons Institute in Fall 2021 to participate in the program Computational Complexity of Statistical Inference.

• In the summer of 2021, I led a Data+ project, see student project presentation here: Detecting and Matching Similar Networks.

Research themes

• Network alignment: Given two noisy networks with unknown node labels, can we tell whether they are related and identify which nodes correspond?
• Flexibility design in matching markets: Before a market forms, a platform chooses how much flexibility to give each agent or each side. How should a limited budget be allocated to enable more matches?
• Hiring & screening: Signals, interviews, and AI-assisted reports change what applicants and employers learn—and which matches or hiring decisions follow.
• Online metric matching: Supply is available from the start. Requests arrive one at a time and must be matched immediately to unused resources, with the aim of minimizing total distance.
• Dynamic matching & allocation: As arrivals, available resources, and budgets change, a system must decide whom to match and what to allocate while preserving future opportunities.
• Hidden network geometry: Humans often connect through shared interests or traits. If bots imitate human accounts, could those patterns of connection still reveal the presence of a human community?

Preprint and Publications

• Pathwise-Budgeted LLM Routing: Online Allocation with K-Bit Tail-Cost Learning

  Hanzhang Qin, Sophie H. Yu, Yuewei Zhang, Zhi-Hai Zhang

  Dynamic matching & allocationExplore the idea ↗

• Screening AI-equipped applicants

  Pengyu Qian and Sophie H. Yu

  Hiring & screeningExplore the idea ↗Play the game ↗

• When do token systems stabilize? Compatibility constraints and community structure

  Flore Sentenac, Mingwei Yang, Sophie H. Yu

  Dynamic matching & allocationExplore the idea ↗

• Optimal analysis of greedy for stochastic online euclidean matching

  Mingwei Yang, Sophie H. Yu

  Online metric matchingExplore the idea ↗

• Distance flexibility in spatial matching: the value of concentration

  Taha Ameen, Sophie H. Yu

  Flexibility designExplore the idea ↗

• Flexibility allocation in random bipartite matching markets: exact matching rates and dominance regimes

  Taha Ameen, Flore Sentenac, Sophie H. Yu

  Flexibility designExplore the idea ↗

• Detection of local geometry in random graphs: information-theoretic and computational limits

  Jinho Bok, Shuangping Li, Sophie H. Yu

  Hidden network geometryExplore the idea ↗

• A uniformity principle for spatial matching

  Taha Ameen, Flore Sentenac, Sophie H. Yu

  Flexibility designExplore the idea ↗

• Optimal regret for dynamic matching with static priorities

  Süleyman Kerimov, Pengyu Qian, Mingwei Yang, Sophie H. Yu

  Major revision at Management Science, 2026

  Dynamic matching & allocationExplore the idea ↗

• From signaling to interviews in random matching markets

  Maxwell Allman, Itai Ashlagi, Amin Saberi, and Sophie H. Yu

  Major revision at Management Science, 2026

  A short version appears at ACM Symposium on Theory of Computing (STOC), 2025

  Hiring & screeningExplore the idea ↗

• Dynamic resource allocation without re-solving: the effectiveness of primal-dual policies

  Siqi He, Yehua Wei, Jiaming Xu and Sophie H. Yu

  Major revision at Operations Research, 2026

  Dynamic matching & allocationExplore the idea ↗

• Constant regret primal-dual policy for multi-way dynamic matching

  Yehua Wei, Jiaming Xu and Sophie H. Yu

  Accepted at Management Science, 2026

  A short version appears at ACM SIGMETRICS, 2023

  Dynamic matching & allocationExplore the idea ↗

• Online metric matching: beyond the worst case

  Mingwei Yang and Sophie H. Yu

  Accepted at Operations Research, 2026

  The conference version is accepted at The 20th Conference on Web and Internet Economics (WINE), 2024

  Online metric matchingExplore the idea ↗

• Random graph matching at Otter's threshold via counting chandeliers

  Cheng Mao, Yihong Wu, Jiaming Xu, and Sophie H. Yu

  Accepted to Operations Research, 2025

  A short version is accepted in ACM Symposium on Theory of Computing (STOC), 2023

  Network alignmentExplore the idea ↗

• Testing network correlation efficiently via counting trees

  Cheng Mao, Yihong Wu, Jiaming Xu, and Sophie H. Yu

  Accepted to Annals of Statistics, 2023

  George Nicholson Student Paper Competition finalist, INFORMS 2022

  Algorithm available here

  Network alignmentExplore the idea ↗

• Testing correlation of unlabeled random graphs

  Yihong Wu, Jiaming Xu, and Sophie H. Yu

  Accepted to Annals of Applied Probability, 2022

  Network alignmentExplore the idea ↗

• Settling the sharp reconstruction thresholds of random graph matching

  Yihong Wu, Jiaming Xu, and Sophie H. Yu

  Accepted to IEEE Transactions on Information Theory, 2022

  A short version appears at IEEE International Symposium on Information Theory (ISIT), 2021

  Network alignmentExplore the idea ↗

• Differentially private verification of regression predictions from synthetic data

  Haoyang Yu and Jerome P. Reiter

  Accepted to Transactions on Data Privacy, 2018

Students

Students I’ve worked closely with:

• Taha Ameen (UIUC, CMU)
• Mingwei Yang (Stanford MS&E)
• Bora Cetin (Wharton OIDD)
• Jinho Bok (Wharton Stats)

Selected Talks

• INFORMS Annual Meeting, Nov, 2026

• Carnegie Mellon University, Operations Seminar, Sep, 2026

• Allerton Conference, UIUC, Sep, 2026

• Renmin University of China: Information (Aug); Business and Gaoling AI (June), 2026

• Tsinghua University, School of Economics and Management, Operations Seminar, June, 2026

• Mostly OM, Tsinghua University, May, 2026

• Hong Kong University of Science and Technology, IEDA Seminar, May, 2026

• The Chinese University of Hong Kong, Operations Management Seminar, May, 2026

• The University of Hong Kong (HKU), Operations Management Seminar, May, 2026

• OM in the Wild Workshop, May, 2026

• Stanford University, Guest lecture, Nov, 2025

• INFORMS Annual Meeting, Oct, 2025

• University of Pennsylvania, Theory of Computer Science Seminar, Oct, 2025

• Allerton Conference, UIUC, Sep, 2025

• INFORMS Applied Probability Society Conference, Jul, 2025

• INFORMS Revenue Management and Pricing, Jul, 2025

• The 57th ACM Symposium on Theory of Computing (STOC), June, 2025

• Marketplace Innovations Workshop, May, 2025

• OM in the Wild Workshop, May, 2025

• University of UT Austin, IEOR Seminar, Jan, 2025

• University of Toronto, Rotman Young Scholar Seminar, Dec, 2024

• University of Pennsylvania, Operations Seminar, Dec, 2024

• INFORMS Annual Meeting, Oct, 2024

• Institut de Mathématiques de Toulouse (IMT), Online Stochastic Matching, Sep, 2024

• Stanford University, Rain Seminar, Apr, 2024

• London School of Business, Operations Management Seminar, Feb, 2024

• Shanghai Jiao Tong University, Antai College of Economics, Jan, 2024

• INFORMS Annual Meeting, Oct, 2023

• INFORMS Applied Probability Society Conference, June, 2023

• ACM SIGMETRICS, June, 2023

• The 55th ACM Symposium on Theory of Computing (STOC), June, 2023

• MIT FODSI Computational Complexity of Statistical Problems Workshop, June, 2023.

• University of California, Davis, Probability Seminar, May, 2023.

• Marketplace Innovations Workshop, May, 2023

• Stanford University, GSB, OIT Seminar, Feb, 2023

• Columbia University, IEOR Seminar, Feb, 2023

• Duke University, Department of Computer Science, CS Theory Seminar, Jan, 2023

• University of Southern California, Marshall, Operations Seminar, Jan, 2023

• University of Southern California, Marshall, Statistics Seminar, Jan, 2023

• Northwestern University, Statistics Seminar, Jan, 2023

• University of Chicago, Booth, Operations Management Seminar, Jan, 2023

• Northwestern University, Kellogg, Operations Seminar, Dec, 2022

• Cornell University, Ithaca, ORIE Seminar, Dec, 2022

• Cornell Tech, ORIE Seminar, Dec, 2022

• Emory University, Goizueta, ISOM Seminar, Dec, 2022

• Georgia Institute of Technology, ISyE Seminar, Dec, 2022

• University of Pennsylvania, Wharton, OID Seminar, Dec, 2022

• University of Southern California, Viterbi School of Engineering, ISE Seminar, Nov, 2022

• University of California, Davis, GSM, Business Analytics Seminar, Nov, 2022

• INFORMS Annual Meeting, Session TB41, Oct, 2022

• INFORMS Annual Meeting, George Nicholson Student Paper Competition, Oct, 2022

• Cornell University, ORIE Young Researchers Workshop, Oct, 2022

• Duke University, Fuqua School of Business, Interdisciplinary Seminar, Jan, 2022

• University of California, Berkeley, Simons Institute, CCSI Student Seminar, Nov, 2021

• INFORMS Annual Meetings, Session WC08, Oct, 2021

• IEEE International Symposium on Information Theory Conference, Jul, 2021

Teaching

• Instructor, EGRMGMT-580 Decision Modeling (Graduate) at Pratt school of engineering, Duke University, Spring 2022

  (Instructor Rating: 4.82/5.00, Response rate: 18/21, Department mean: 4.38/5.00)

• Teaching Assistant, PhD BA 915/ STA 715/ MATH 742: Stochastic Models, Duke University, Fall 2022

• Teaching Assistant, Weekend MBA Decision 611: Decision Models, Duke University, Fall 2022

• Teaching Assistant, PhD BA 990/ ECE 590: Statistical Inference on Graphs, Duke University, Spring 2022

• Teaching Assistant, PhD BA 910/Statistics 502: Bayesian Inference, Duke University, Spring 2021

• Teaching Assistant, MBA Decision 516: Quantitative Business Analysis, Duke University, Fall 2020

• Teaching Assistant, MBA Decision 521Q: Quantitative Business Analysis, Duke University, Fall 2020

• Teaching Assistant, MQM Decision 518Q: Applied Probability and Statistics, Duke University, Fall 2019

• Teaching Assistant, MQM Decision 521Q: Decision Analytics and Modeling, Duke University, Spring 2019

• Teaching Assistant, Economics 618: Advanced Econometrics, Duke University, Fall 2016