Welcome to my homepage!

My name is Parshan Pakiman. I am a Ph.D. candidate in Information and Decision Sciences (IDS) at the University of Illinois Chicago (UIC). I am grateful for being co-advised by Selva Nadarajah and Negar Soheili. Before joining UIC, I received my Bachelor of Science in Mathematics and Applications from the School of Mathematics, Statistics, and Computer Science at the University of Tehran.

My main research interests include:

(i) Off-the-shelf reinforcement learning (RL) algorithms: Mitigating the burden of model selection and parameter hand-engineering to broaden the use of RL in business applications (i.e., dynamic pricing with demand learning, options pricing, marketing campaign optimization, inventory control) and making it accessible to non-experts.


(ii) Learning from sequential decisions: Uncovering unknown parameters of an optimization problem used to make historical decisions via inverse RL and then using learned parameters in a forward RL to enhance past decisions.


(iii) Technical expertise: Advancing the above themes by developing methods and theory based on approximate linear programming, random features, information relaxations and duality, and online convex programs.

July 25-28, 2022: Looking forward to joining the ICCOPT 2022 conference at Lehigh University and presenting my work!

October 16, 2022: Very excited to co-organize a session with professor Selva Nadarajah on Recent Advances in Sequential Decision Making at the 2022 INFORMS Annual Meeting. We have an amazing set of speakers, professors Canan Ulu, Michael Pavlin, and Peter Frazier.

April 24, 2022: I will give two talks at the 32nd Annual POMS conference. The title of my presentations are Balancing Financial and Social Objectives via Decision Learning and Menu Optimization and Self-adapting Reinforcement Learning for Financial and Real Options.

March 13th, 2022: I am co-organizing a session with Professor Selva Nadarajah at the 2022 INFORMS Optimization Society, conference, which will be held at Greenville, SC. The session will focus on linear programming approximations for large-scale applications. The list of talks will be announced soon!

October 24th, 2021: At INFORMS 2021, I will chair the Recent Advances in Reinforcement Learning session, which is part of the Optimization under Uncertainty cluster. I will also co-chair Social Responsibility and Risk in Supply Chains session under the Social Responsibility and Risk in Supply Chains cluster with Dr. Sanjith Gopalakrishnan.

May 1st, 2021: I will give a talk at the Environmental Sustainability and Social Responsibility session of POMS 2021. The title of my presentation is Putting Social Responsibility on the Menu: AI-Guided Tool Selection that Aligns Worker and Social Objectives.

December 12th, 2020: Our paper titled Self-guided Approximate Linear Programs has been accepted at the Workshop on Self-Supervised Learning Theory and Practice, NeurIPS 2020. self-adapting robustness in demand learning. Link to the paper.

November 10th, 2020: At the 2020 INFORMS Annual Meeting, I will have a presentation under the MSOM's supply chain cluster with the title of self-adapting robustness in demand learning. Link to the Advances in Supply Chain Optimization track.

Sep 22nd, 2020: I presented a paper entitled self-adapting robustness in demand learning at INFORMS revenue management and pricing (RMP) student live paper series. Link to my presentation recording.

April 27th, 2020: I had a short presentation regarding my ongoing project with the title of managing packing efficiency and sustainability in e-commerce: a semi-supervised learning approach at the Symposium on Energy, Environment & Sustainability (SEES, April 2020). Schedule , booklet , and webinar recording.

(Canceled due to COVID-19) March 15, 2020: I chair a session with the title of Advances in Approximate Dynamic Programming and Reinforcement Learning at the INFORMS Optimization Society Conference (IOS2020) . The titles of presentations at this session are Lookahead-bounded Q-learning, Network-based Approximate Linear Programming, and Self-guided Approximate Linear Programs.

October 22nd, 2019: Professor Selva Nadarajah presents our joint work at INFORMS Annual Meeting 2019. The title of our paper is self guided approximate linear programs and it is scheduled under the learning algorithms (theory and applications) track. See schedule.

August 8th, 2019: I give a talk in the 25th ACM SIGKDD conference on knowledge discovery and data mining. See schedule.