Hello and thanks for visiting my page.
My name is Sina Aghaei and I am a postdoctoral fellow at Harvard Kennedy School.
I work at the intersection of machine learning, optimization and causal inference where I design robust, interpretable and fair machine learning and prescriptive models.
I earned my PhD in Operations Research from University of Southern California (USC) under the guidance of Phebe Vayanos. Prior, I completed my Master's in Computer Science at USC. My undergraduate studies included a double major in Computer Engineering and Industrial Engineering at Sharif University of Technology.
News
March 2024: I'm honored to have been named a Rising Star in Management Science & Engineering at Stanford.
January 2024: I joined the Public Impact Analytics Science Lab (PIAS-Lab) at Harvard Kennedy School as a postdoctoral fellow, collaborating with Prof. Soroush Saghafian.
April 2024: I'm taking a course on AI Governance.
November 2023: I won the 3rd place prize in 3 Minute Thesis (3MT) competition at USC. Check out the event and my presentation here.
October 2023: Our paper, 'Strong Optimal Classification Trees,' has been accepted for publication in Operations Research.
May 2023: Check out the photos from my PhD hooding ceremony! (Photo album link, Full ceremony video link)
May 2023: Our paper "Learning Optimal Fair Classification Trees" got accepted in the 6th AAAI/ACM Conference on AI, Ethics, and Society (AIES), 2023.
December 2022: Our paper "Fairness in Contextual Resource Allocation Systems: Metrics and Incompatibility Results" got accepted in the 37th AAAI Conference on Artificial Intelligence, 2023.
April 2022: I am co-organizing a session on "Interpretability, Fairness and Robustness in OR/ML" this year at INFORMS Annual Meeting 2022, Indianapolis. [Click here for the session info]
Aug 2022: My internship project at Snap won the 1st place in DXP hackathon.
May 2022: I got a student scholarship to attend CPAIOR 2022.
April 2022: I'll give a guest lecture at the University of Michigan on "Machine Learning and Integer Optimization for Analytics in High-Stakes Domains".
Janurary 2022: I'll join Snapchat as a software engineering intern for summer 2022!
January 2022: I'll be the teaching assistant for the course "Analytics for Social Impact" instructed by Phebe Vayanos in Spring 2022 semester at USC.
December 2021: Our paper "Learning optimal prescriptive trees from observational data" got accepted in AAAI Workshop on AI for Behavior Change, 2022.
December 2021: Our paper "Optimal Robust Classification Trees" got accepted in AAAI Workshop on Adversarial Machine Learning and Beyond, 2022.
Januray 2022: I will serve as the chair of "Machine learning methods" session at 2022 INFORMS COMPUTING SOCIETY (ICS) CONFERENCE , Tampa, Florida.
October 2021: I presented "Learning Fair Optimal Trees" at INFORMS annual meeting, Anaheim, California.
July 2021: I presented "Learning Fair Optimal Trees" at 31st Europian Conference on Operational Research (Euro 2021)
April 2021: I presented a poster "Learning Fair Optimal Trees" at 1st Equity and AI CAIS Symposium.
March 2021: I passed my candidacy exam! I am now a PhD candidate!
June 2021: I will be talking about "Fair and Strong Optimal Decision Trees" at "Socially Responsible AI" mini-syposium at SIAM's Conference on Financial Mathematics and Engineering.
November 2020: I will be a mentor in a STEM outreach program to raise high school students' interest in pursuing STEM fields. Check out more details here. You can find my slides here!
November 2020: I am co-organizing a session on "Fairness in Machine Learning and Optimization" this year at INFORMS Annual Meeting 2020.
September 2020: I will be giving a talk on "Learning optimal decision trees" at CPAIOR 2020.
August 2020: I will be giving a talk on "Learning optimal decision trees" at Amirkabir's Summer Summit in Artificial Intelligence 2020.
May 2020: My poster "Learning optimal decision trees: strong max-flow formulations" got accrepted in MIP2020 workshop.
May 2020: I will be presenting our work on "Learning optimal decision trees: strong max-flow formulations" at INFORMS annual meeting, National Harbor, Maryland.