Completed and Ongoing Research

Optimal Ethical Decision-Making

• Differential privacy via distributionally robust optimization
  AS, Huikang Liu and Wolfram Wiesemann
  Operations Research (2025)
  [preprint]
  ✨ George Nicholson Student Paper Competition (2024; second place)
✨ Imperial Operations and Analytics Best PhD Paper Award (2023; first place)
✨ INFORMS Optimization Society Student Paper Prize (2023; honorable mention)

• Room for improvement: Optimal and fair housing allocation via weakly coupled Markov decision processes
  AS, Bill Tang, Cagil Kocyigit, Phebe Vayanos, and Wolfram Wiesemann
  Working Paper
  

• Mind the Gap: Mixtures of Gaussians in approximate differential privacy
  Huikang Liu, AS, and Wolfram Wiesemann
  International Conference on Machine Learning (ICML 2026)
  ☆ Appeared: INFORMS Workshop on Data Science (2025)
☆ Appeared: INFORMS Workshop on Data Mining and Decision Analytics (2025)
  

Robust Machine Learning

• It's all in the mix: Wasserstein classification and regression with mixed features
  Reza Belbasi, AS and Wolfram Wiesemann
  Manufacturing & Service Operations Management (2026)
  [preprint]

• Distributionally and adversarially robust logistic regression via intersecting Wasserstein balls
  AS, Eleonora Kreacic, Mohsen Ghassemi, Vamsi Potluru, Tucker Balch and Manuela Veloso
  Uncertainty in Artificial Intelligence (UAI 2025)
  [preprint]
  ✨ Oral Presentation
✨ INFORMS Data Mining Society Best Student Paper Competition (2024; finalist)
☆ Appeared: INFORMS Workshop on Data Science (2024)
  

• Wasserstein logistic regression with mixed features
  AS, Reza Belbasi, Martin Haugh and Wolfram Wiesemann
  Neural Information Processing Systems (NeurIPS 2022)
  [preprint]

Nonconvex Optimization

• Maximizing the sum of maxima of convex functions
  AS, Aharon Ben-Tal, and Wolfram Wiesemann
  In preparation for submission

• Convex maximization via adjustable robust optimization
  AS, Aharon Ben-Tal, Ruud Brekelmans and Dick den Hertog
  INFORMS Journal on Computing (2022)
  [preprint]
  ✨ INFORMS Computing Society Student Paper Award (2021; first place)

• A reformulation-linearization technique for optimization over simplices
  AS, Dick den Hertog and Wolfram Wiesemann
  Mathematical Programming [Short Communication] (2023)
  [preprint]

• Using column generation to solve extensions to the Markowitz model
  Lorenz M. Roebers, AS and Juan Vera
  The Engineering Economist [Technical Note] (2019)
  [preprint]

Doctoral Dissertation

• Models and algorithms for safeguarded data-driven decision making
  AS [Imperial Spiral]
  ✨ George Dantzig Dissertation Award (2025; finalist)
✨ G-Research Imperial College PhD competition (2024; first place)