Lin An
I am a fifth-year PhD candidate in the Algorithms, Combinatorics, and Optimization program at Carnegie Mellon University’s Tepper School of Business. My research focuses on leveraging AI to enhance decision-making through the integration of optimization and machine learning. I study stochastic, online, and prediction-based models. My work is motivated by operations management applications, including resource allocation, personalization systems, and inventory management.
Education
Ph.D., Algorithms, Combinatorics, and Optimization, Carnegie Mellon University, 2026
M.S., Operations Research, Columbia University, 2021
B.S., Mathematics and Economics, Boston College, 2020
Research Themes
Optimization for AI-driven models
I study how modern AI architectures, such as transformers, can be combined with optimization to support real-time decision-making. My work focuses on designing algorithms that retain the modeling power of AI while ensuring computational tractability, scalability, and provable guarantees.
Decisions with predictions
I develop algorithms that incorporate predictions generated by machine learning models into decision-making. These methods exploit predictive information when it is accurate while maintaining robust performance when predictions are imperfect, allowing data-driven decisions without sacrificing reliability.
Algorithms for sequential decision-making
I study stochastic, online, and dynamic optimization models motivated by operations management applications. My work emphasizes algorithmic design with provable guarantees while maintaining practical performance in sequential and real-time settings.
Papers and Work in Progress
-
Sequential Engagement Prediction with Interpretable AttentionIn preparation · 2026
-
Efficient Attention-Based Personalization via Clustering of Contextual StatesIn preparation · 2026
-
In preparation · 2026A preliminary version appeared in Marketing Dynamics Conference (MDC 2026).
-
Management Science, Submitted · 2025A preliminary version appeared in NeurIPS ML×OR Workshop (NeurIPS 2025).A preliminary version appeared in ACM SIGMETRICS Performance Evaluation Review (SIGMETRICS 2025).
-
Management Science, Minor revision · 2024A preliminary version appeared in the refereed INFORMS Optimization Society Conference (IOS 2024).
-
Manufacturing & Service Operations Management, Published · 2023
-
Proceedings of the ACM-SIAM Symposium on Discrete Algorithms (SODA), Published · 2022
-
PhD Dissertation · 2026
Awards and Honors
Finalist, INFORMS Data Mining and Decision Analytics Best Paper Competition, 2025
GenAI Fellowship, CMU Center for Intelligent Business, 2024
Egon Balas Award, Tepper School of Business, 2023
William Larimer Mellon Fellowship, Carnegie Mellon University, 2021
Teaching
Mathematical Models for Consulting (Undergraduate) · Instructor
Carnegie Mellon University, Fall 2024
Course evaluations: 4.92 / 5.00
Operations Strategy (MBA) · Teaching Assistant
Carnegie Mellon University, Fall 2023
Course evaluations: 4.85 / 5.00
Optimization (Undergraduate) · Teaching Assistant
Carnegie Mellon University, Fall 2023
Course evaluations: 4.78 / 5.00
Advanced Graph Theory (PhD) · Teaching Assistant
Carnegie Mellon University, Fall 2022
Course evaluations: 5.00 / 5.00
Service
Reviewer
Management Science; ACM Conference on Economics and Computation (EC); Operations Research Letters
Program Committee Member
INFORMS Workshop on Data Science, 2024, 2025
Session Chair
INFORMS Optimization Society Conference, 2026
Revenue Management and Pricing Track, POMS Annual Conference, 2025
Speakers Chair
YinzOR, 2023, 2024, 2025
Vice President
CMU INFORMS Student Chapter, 2023, 2024
Collaborators
I welcome collaboration across academia and industry. My work sits at the intersection of AI, optimization, and operations management, and I am especially interested in projects that combine mathematical structure with impactful real‑world applications.
Prospective PhD Students
I am particularly interested in working with PhD students in business schools whose research interests overlap with AI, optimization, and operations management. If you are a current PhD student and are interested in discussing potential research collaborations, feel free to reach out.
Students with strong backgrounds in mathematics, computer science, engineering, or related quantitative areas are especially welcome. For students coming from other fields, it may be useful to know that business schools offer opportunities to work on rigorous mathematical and algorithmic problems while also engaging with important real‑world applications.
Undergraduate and Master's Students
I supervise research projects with undergraduate and master's students interested in AI, optimization, and data‑driven decision‑making. These projects often involve mathematical modeling, algorithm design, or empirical analysis motivated by real applications.
Students who work with me frequently use these experiences as preparation for PhD programs or quantitative careers in industry. Research experience can help students gain admission to strong PhD programs or obtain positions in data science, technology, and related fields.
If you are interested in research, please email me with a brief description of your background, relevant coursework, and the types of problems that interest you.
Academic Collaborators
I am always interested in discussing research ideas with faculty and researchers working in optimization, machine learning, operations management, or related areas. If you have a problem that intersects with my interests, feel free to reach out.
Industry Collaborators
Many of my research projects are motivated by practical decision problems. I am happy to collaborate with industry partners who have interesting datasets or challenging operational problems in areas such as personalization systems, resource allocation, or data‑driven decision‑making.