Faraz Khoshbakhtian

scholar github twitter linkedin

About

I am an Industrial Engineering PhD candidate at University of Toronto. My research focuses on the intersection of machine learning and combinatorial optimization, with particular emphasis on applications in healthcare and pandemic mitigation strategies. I develop novel learning algorithms that can be used to solve large-scale optimization problems such as the Critical Node Detection Problem.

I am a member of morLAB and I am advised by Prof. Dionne M. Aleman.

I am currently an applied science intern at Amazon science where I work on novel network representation learning techniques and applications in supply chain optimization. I was previously at Mastercard AI Garage, where I developed new (multi-task) self-supervised learning algorithms for generalizable node embeddings on large-scale graphs. As you may have guessed, I am very passionate about the potential of graph learning in real-world applications.

Beyond the realm of engineering, my academic foundation is rooted in Philosophy from my undergraduate studies, and I continue to regard myself as an ongoing student in this discipline. Outside of research, I enjoy making music, photography, wandering on long walks, and cinema .

My resume is available here.

Research

I am deeply curious about the intricate dynamics of influence and power in complex systems. Right now, my research is zoomed in on large-scale pandemic modeling, especially where network representation learning meets the Critical Node Detection Problem (CNDP).

In my MASc program during 2020-21, I worked on some cool machine learning applications, mainly focusing on the early prognosis of COVID-19 patients. This experience gave me a solid foundation, propelling me into my current PhD journey.

I earned my HBSc in Philosophy, Computer Science, and Statistics from the University of Toronto in 2020. Along the way, I picked up a couple of awards—Woodsworth College Brookfield’s Leadership Scholarship in 2018 and the Scholars in Residence award in 2017. I also had the chance to be part of the Socrates project at the Philosophy Department in 2018, which was an incredible experience.

Academic writings and publications

  • Khoshbakhtian, F., Gaurav, O., Aleman, A., and Asthana, S. (2024). MEGA: Multi-Encoder GNN Architecture for stronger task collaboration and generalization. accepted for publication at Lecture Notes in Computer Science Series (LNCS). (preprint, code)
  • Khoshbakhtian, F., Validi, H., Ventresca, M., Aleman, D. M. (2023) Distance-based critical node detection for effective vaccine policies. accepted for publication at Operations Research Letters. (preprint, code)
  • Pirmorad, E., Khoshbakhtian, F., Mansouri, F., and Farahmand, A. M. (2021) Deep reinforcement learning for online control of stochastic partial differential equations. The Symbiosis of Deep Learning and Differential Equations. (pdf)
  • Khoshbakhtian, F., Lagman, A., Aleman, D. M., Giffen, R., and Rahman, P. (2021). Prediction of severe COVID-19 infection at the time of testing: A machine learning approach. (preprint)
  • Khan, S. S., Khoshbakhtian, F., and Ashraf, A. B. (2021). Anomaly detection approach to identify early cases in a pandemic using chest X-rays. Proceedings of the Canadian Conference on Artificial Intelligence. (pdf)

Conference and workshop presentations

(bold for the presenter)

  • Khoshbakhtian, F., Ahamd, A., Cohen, A., and Aleman, D. M. Enhancing critical node detection with beam search: a heuristic-agnostic approach. INFORMS Annual Meeting. Seattle, US. October 2024 (scheduled).
  • Khoshbakhtian, F., Gaurav, O., Aleman, A., and Asthana, S. MEGA: Multi-Encoder GNN Architecture for stronger task collaboration and generalization. ECML PKDD. Vilnius, Lithuania. September 2024 (scheduled).
  • Khoshbakhtian, F., Ahamd, A., Cohen, A., and Aleman, D. M. Enhancing critical node detection with beam search: a heuristic-agnostic approach. CORS Annual Meeting. London, Canada. June 2024.
  • Khoshbakhtian, F., Validi, H., Ventresca, M., Aleman, D. M. Distance-based critical node detection for effective vaccine policies. INFORMS Annual Meeting. Phoenix, US. October 2023 (scheduled)
  • Khoshbakhtian, F., Validi, H., Ventresca, M., Aleman, D. M. Distance-based critical node detection for effective vaccine policies. INFORMS Healthcare. Toronto, Canada. July 2023
  • Khoshbakhtian, F., Validi, H., Ventresca, M., Aleman, D. M. Distance-based critical node detection for effective vaccine policies. CORS / Optimization Days. Montreal, Canada. May 2023.
  • Khoshbakhtian, F., Validi, H., Ventresca, M., Aleman, D. M. Distance-based critical node detection for effective vaccine policies. Panoptic: view on global optimization. Florida, US. March 2023.
  • Khoshbakhtian, F., Lagman, A., Aleman, D. M., Giffen, R., and Rahman, P. Prediction of severe COVID-19 infection at the time of testing: A machine learning approach. CORS/INFORMS International Conference. Vancouver, Canada. June 2022.
  • Pirmorad, E., Khoshbakhtian, F., Mansouri, F., and Farahmand, A. M. Deep reinforcement learning for online control of stochastic partial differential equations. Spotlight presentation at The Symbiosis of Deep Learning and Differential Equations. virtual. Dec 2021.

Other writings

  • Stories of Autonomy: A Critical Introduction in Defense of a Belief in the Indifference of the Will (pdf)

Teaching

At the university of Toronto, I have served as a Teaching Assistant and Head Teaching Assistant for a series of programming, data analytics, and philosophy courses:

  • MIE1624: Introduction to Data Science and Analytics
  • MIE253: Data Modelling
  • MIE250: Introduction to Object Oriented Programming
  • APS106: Fundamentals of Computer Programming
  • PHL101: Introduction to Philosophy

News

  • July 2024: Relocated to Vancouver and started an applied science internship at Amazon Science
  • June 2024: Distance-based critical node detection for effective vaccine policies got accepted for publication at Operations Research Letters
  • June 2024: MEGA: Multi-Encoder GNN Architecture for stronger task collaboration and generalization got accepted for a conference presentation at PKDD2024 Research Track
  • March 2024: Extension of the Mastercard internship
  • August 2023: New personal website is now live!
  • August 2023: Won Ontario Gratuade Scholarship (OGS) (value: 15,000CAD)
  • June 2023: Started an applied science internship at Mastercard
  • May 2023: Won Mitacs Accelerate award (value: 15,000CAD)
  • April 2023: Won Emerging and Pandemic Infections Consortium doctoral award (value: 10,000CAD)
  • October 2022: Co-founded University of Toronto Students for a Free Iran (UTSFI)