Conference Papers
- Conformal Inverse Optimization
(with Timothy Chan and Erick Delage)
NeurIPS'24 (acceptance rate: 25.8%)
[Open Review] , [PDF], [5-min Talk], [3-min AI Podcast Intro] - AutoLTS: Assessing Cycling Stress via Contrastive Learning and Spatial Post-processing
(with Timothy Chan and Shoshanna Saxe)
AAAI'24 (acceptance rate: 24.9%)
[DOI], [PDF], [Code & Data]
- Preliminary version accepted to AAAI'23 Workshop on AI for Social Good
Journal Papers
- Machine Learning-Augmented Optimization of Large Bilevel and Two-stage Stochastic Programs: Application to Cycling Network Design
(with Timothy Chan and Shoshanna Saxe)
Major revision, M&SOM, 2024
[arXiv], [PDF], [Code], [5-min Talk] [12-min AI Podcast]
- Implemented to support Toronto's 2025-2030 bike infrastructure planning
- Winner, INFORMS 2024 TSL Best Student Paper Award
- Winner, NUS Analytics for X 2024 Best Student Presentation Award
- Runner Up, CORS 2023 Best Student Paper Competition
- Finalist, INFORMS 2024 Service Science Best Cluster Paper Award
- Finalist, 2024 Gilbert Laporte Student Paper Award
- Media Coverage: University of Toronto News, TechXplore - Conformal Inverse Optimization for Adherence-aware Prescriptive Analytics
(with Timothy Chan and Erick Delage)
Under review, 2024
Preliminary: NeurIPS'24
[SSRN], [Short Version], [Full Version], [5-min Talk], [60-min Talk], [3-min AI Podcast Intro]
- Finalist, INFORMS 2024 DMDA Workshop Best Paper Award - Theoretical Track - Exploring the Geographical Equity-Efficiency Trade-off in Cycling Infrastructure Planning
(with Madeleine Bonsma-Fisher, Timothy Chan and Shoshanna Saxe)
Journal of Transport Geography, 2024
[DOI], [SSRN], [PDF]
- Media Coverage: University of Toronto News, TechXplore - The Impact of COVID-19 Cycling Infrastructure on Low-Stress Cycling Accessibility: A Case Study in the City of Toronto
(with Timothy Chan and Shoshanna Saxe)
Findings, 2021
[DOI], [PDF], [Data]
- Media Coverage: University of Toronto News, CBC News, Toronto Star - Deep Reinforcement Learning for Electric Vehicle Routing with Time Windows
(with Bissan Ghaddar and Jatin Nathwani)
IEEE Transactions on Intelligent Transportation Systems, 2022
[DOI], [arXiv], [PDF] - Optimizing Electric Vehicle Routing and Charging/Discharging Under Time-Variant Electricity Prices
(with Bissan Ghaddar and Jatin Nathwani)
Transportation Research Part C: Emerging Technologies, 2021
[DOI], [arXiv], [PDF] - Text Mining Applications to Undergraduate Engineering Programs: Analyzing Differences in Gender, Nationality and Socio-economic Status
(with Bissan Ghaddar and Ada Hurst)
Likely a permanent working paper, 2021
[arXiv], [PDF]
Work in Progress
- Analytics for Better Urban Cycling: A Multi-year Collaboration with the City of Toronto
(with Madeleine Bonsma-Fisher, Timothy Chan and Shoshanna Saxe)
- Finalist, CORS 2025 Practice Prize (Winner TBD) - Contextual Risk Measures
(with Timothy Chan and Erick Delage) - Bridging Predictive and Prescriptive Analytics in Urban Cycling Network Design
(with Timothy Chan and Shoshanna Saxe)
Thesis
Analytics for Better Urban Cycling
(advised by Timothy Chan and Shoshanna Saxe)
PhD thesis, University if Toronto, March 2025
Open Source
- AutoLTS: Computer vision models to assess cycling stress of road networks:
- LTS-TRT: The level of traffic stress network in the City of Toronto (July 2021):
- R2D-VRP: A hierarchical reinforcement learning framework for solving the vehicle routing problem :
- OCT: Python implementations of optimal classification tree formulations and their computational performance:
- CycleLinx: An interactive web app for exploring Toronto’s bicycle network and our optimized expansion plan:
Conference Presentations
- INFORMS: 2020, 2022, 2023, 2024
- INFORMS ICS: 2025
- CORS: 2019, 2021, 2022, 2023, 2024
- AAAI: 2023, 2024
- Vector Research Symposium: 2023
- TRB: 2024
- POMS-HK: 2024
- MSOM: 2024
- ISMP: 2024
- NeurIPS: 2024