Ruihao Zhu
 |
Candidate for Interdisciplinary Ph.D. in Statistics (IDPS) Contact77 Massachusetts Ave, Bldg E18-436, Cambridge, MA 02139 |
About
Ruihao Zhu is currently a fifth year candidate for the Interdisciplinary Ph.D. in Statistics (IDPS) at the MIT Institute for Data, Systems, and Society (IDSS) and Laboratory for Information & Decision Systems (LIDS). He has the pleasure of being advised by David Simchi-Levi. Born in Guangzhou, China, Ruihao received his B.Eng. degrees in Electrical Engineering and Computer Science from both Shanghai Jiao Tong University and the University of Michigan in 2015. He spent the summers of 2020 and 2019 as research scientist intern with Amazon Supply Chain Optimization Technologies (SCOT) and Google Research Mountain View.
Research
Ruihao's research seeks to help organizations to improve decision-making in uncertain and dynamic environments by developing data-driven models and analytics methodologies. He is currently focusing primarily on reinforcement learning, transfer learning, and high-dimensional statistics, with applications in business analytics, online marketplaces, fairness & privacy, and supply chain management. As part of it, he enjoys collaborting with companies across different industries, such as automobile, consumer goods, and healthcare. Some of the projects he has worked on include
• Non-stationary reinforcement learning and revenue management in dynamic environments • Designing dynamic pricing experiments across related products • Robust sales forecast calibration amid the COVID-19 pandemic (in collaboration with AB InBev) [more details] [paper]
Previously, he has also explored topics related to approximation algroithms, algorithmic game theory, and privacy in service operations.
Journal Publications
Hedging the Drift: Learning to Optimize under Non-Stationarity (with Wang Chi Cheung and David Simchi-Levi)
◇ Forthcoming at Management Science—Special Issue on Data-Driven Prescriptive Analytics.
◇ Preliminary version appeared in Proceedings of the 22nd International Conference on Artificial Intelligence and Statistics (AISTATS 2019).
◇ Honorable Mention, INFORMS George E. Nicholson 2019 Student Paper Competition
◇ Finalist, POMS-JD.com 2019 Best Data-Driven Research Paper Competition
◇ Service Operations SIG Meeting, MSOM Conference 2019
Papers Under Review and Selected Working Papers
Meta Dynamic Pricing: Transfer Learning Across Experiments (with Hamsa Bastani and David Simchi-Levi)
◇ Minor revision at Management Science—Special Issue on Data-Driven Prescriptive Analytics.
◇ Spotlight Track, INFORMS 2019 RM&P Conference
Non-Stationary Reinforcement Learning: The Blessing of (More) Optimism (with Wang Chi Cheung and David Simchi-Levi)
◇ Major revision at Management Science.
◇ Preliminary version appeared in Proceedings of the 37th International Conference on Machine Learning (ICML 2020).
Calibrating Sales Forecast in a Pandemic Using Online Non-Parametric Regression Model (with David Simchi-Levi, Rui Sun, and Michelle X. Wu)
◇ Under review at Management Science.
◇ In collaboration with AB InBev, see here for a summary
Joint Patient Selection and Scheduling under No-Shows: Theory and Application in Proton Therapy (with Soroush Saghafian, Nikolaos Trichakis, and Helen Shih)
◇ Under review at Manufacturing & Service Operations Management.
◇ In collaboration with Massachusetts General Hospital
Near-Optimal Regret Bounds for Model-Free RL in Non-Stationary Episodic MDPs (with Weichao Mao, Kaiqing Zhang, David Simchi-Levi, and Tamer Basar)
◇ Working paper.
Learning to Route Efficiently with End-to-End Feedback: The Value of Networked Structure (with Eytan Modiano)
◇ Working paper.
Selected Conference Publications
Reinforcement Learning for Non-Stationary Markov Decision Processes: The Blessing of (More) Optimism (with Wang Chi Cheung and David Simchi-Levi)
◇ Proceedings of the 37th International Conference on Machine Learning (ICML 2020).
Learning to Optimize under Non-Stationarity (with Wang Chi Cheung and David Simchi-Levi)
◇ Proceedings of the 22nd International Conference on Artificial Intelligence and Statistics (AISTATS 2019).
Coresets for Differentially Private K-Means Clustering and Applications to Privacy in Mobile Sensor Networks (with Dan Feldman, Chongyuan Xiang, and Daniela Rus)
◇ Proceedings of the 26th International Conference on Information Processing in Sensor Networks (IPSN 2017).
Threshold Bandits, With and Without Censored Feedback (with Jacob Abernethy and Kareem Amin)
◇ Advances in Neural Information Processing Systems 29 (NIPS 2016).
Differentially Private and Strategy-Proof Spectrum Auction with Approximate Revenue Maximization (with Kang G. Shin)
◇ Proceedings of the 2015 IEEE International Conference on Computer Communications (INFOCOM 2015).
Differentially Private Spectrum Auction with Approximate Revenue Maximization (with Zhijing Li, Fan Wu, Kang G. Shin, and Guihai Chen)
◇ Proceedings of the 15th ACM International Symposium on Mobile Ad Hoc Networking and Computing (MobiHoc 2014).
Teaching
• MIT Sloan 15.774 The Analytics of Operations Management (Fall 2019)
— Enrollment: 61 (primarily Full-Time MBA, Master of Supply Chain Management, and Master of Business Analytics programs)
— Core requirement for MIT Sloan Business Analytics Certificate
— Rating: 6.0/7.0
— Teaching Assistant for Negin Golrezaei
• MIT 1.S982 Statistical Learning in Operations (Spring 2020)
— Enrollment: 15 (primarily PhD student)
— Teaching Assistant for David Simchi-Levi
Awards
• Honorable Mention, INFORMS George E. Nicholson 2019 Student Paper Competition
• Finalist, POMS-JD.com 2019 Best Data-Driven Research Paper Competition
Professional Services
• Reviewer for Management Science, Operations Research, MSOM Service Operations SIG 2020, Journal of Machine Learning Research (JMLR), International Conference on Machine Learning (ICML) 2020, Conference on Neural Information Processing Systems (NeurIPS) 2019, 2020, International Conference on Algorithmic Learning Theory (ALT) 2019.
• Co-organizer (together with David Simchi-Levi) of MIT Data Science Lab 2019 Workshop on Learning and Optimizing in Operations. [link]
• Coordinator of MIT Data Science Lab seminar series (2019 - now).
Invited Talks
• Amazon Research, Forecasting Team Bandit Workshop (08/2020)
• Kellogg-Wharton OM Workshop (07/2020)
• MIT Data Science Lab Workshop on Learning and Optimizing in Operations (12/2019)
• University of California, Berkeley, Department of Industrial Engineering and Operations Research (IEOR) Summer Seminar Series (06/2019)
• Google AI, Modeling Decisions for Activity-based, Temporal, and Sequential Data (MuDcATS) Weekly Seminar Series (06/2019)
• MSOM, Service Management SIG (06/2019)
• INFORMS RM&P Section Conference, Spotlight Session (06/2019)
• New York University Leonard N. Stern School of Business, Monday Operations Management (MOILS) Seminar Series (04/2019)
• National University of Singapore, Institute of Operations Research and Analytics (IORA) Seminar Series (08/2018)
• Harvard John A. Paulson School of Engineering and Applied Sciences, Economics and Computer Science (EconCS) Seminar Series (03/2017)