Hi there, I'm Afshin Oroojlooy!

I am a principal applied scientists at Oracle OCI where I focused mostly on applications of reinforcement learning and operations research to solve large scale real-world problems. Before joining Oracle, I was a senior RL research scientist at SAS Institute where I spend about 5 years to learn how to build production level machine learning solutions.

My current research is focused on the extension of Reinforcement Learning algorithms for real-world problems, e.g. inventory optimization on multi-echelon networks, traveling salesman problem, stochastic vendor route problem, customer journey optimization, Traffic signal control, HVAC, manufacturing control problems, etc. You can see details of them in my publications or read some overview on my Blog...

My journey on reinforcement learning started by on the industrial engineering department at Lehigh University, where I had the pleasure to work under the supervision of Professor Larry Snyder and Professor Martin Takac. The goal of my dissertation was to extend machine learning and specially reinforcement learning algorithms to solve combinatorial optimization problems in supply chain problems. I got my M.S. of Industrial Engineering from Sharif University of Technology and B.S of Industrial Engineering from Isfahan University of Technology

. Please find my resume Here (updated on 07/10/2021).

Updates:

06/22/2022 Our review paper on multi-agent cooperative RL is got online on Applied Intelligence >
06/22/2022 I joined to Oracle OCI as a principal applied scientists!
04/02/2021 I am presenting " An Introduction to RL and its Applications in Supply Chain Management" at a side workshop at IIISE online conference!.
11/08/2020 I am attending INFORMS Annual Meeting 2020 conference, Online!. I am presenting our new work on traffic signal processing and a review on one section of our MARL review papers, and co-presenting our another presentation on MARL review. INFORMS20
09/27/2020 Our work on application of RL for a traffic signal control accepted on NeurIPS 2020, Vancouver, CA. See the paper [here] NeurIPS
06/25/2020 The beer game paper is accepted on Manufacturing and Service Operations Management (MSOM). Read more about it at blog . We also released the code of paper at github MSOM
08/11/2019 Our review paper on multi-agent cooperative RL is out !.
06/12/2019 I am attending ICML 2019 conference and workshop!. ICML
06/04/2019 The newsvendor paper is accepted on Institute of Industrial and Systems Engineers (IISE) Transaction. Read more about it at blog . We also released the code of paper at github. IISE
11/04/2018 I am attending INFORMS Annual Meeting 2018 conference at Phoenix, AZ!. I am presenting our Beer-Game paper, and co-presenting our VRP paper. INFORMS18
09/16/2018 I Started my full time position as a RL research Scientist at SAS Institute . SAS Institute
09/13/2018 I successfully defended! my dissertation at the ISE Department of the Lehigh University. Lehigh University
09/04/2018 Our RL for VRP paper is accepted at NeurIPS 2018 .! See the code [here] NIPS
03/09/2018 Opex Analytics released an online free version of beer game which uses our AI engine. Try it to see if you can beat the AI! Opex Analytics Beer Game
08/10/2018 I am attending Modeling and Optimization: Theory and Applications (MOPTA) 2018 conference at Bethlehem, PA!. I am presenting our RL for beer game paper and I am co-presenting our RL for VRP paper. MOPTA2018
10/22/2017 I am attending INFORMS Annual Meeting 2017 conference at Houston, AZ!. I am presenting some initial results on our Beer-Game paper. INFORMS17
08/16/2017 I am attending Modeling and Optimization: Theory and Applications (MOPTA) 2017 conference at Bethlehem, PA!. I am presenting our Newsvendor paper and multi-echelon stockout prediction. MOPTA2017
06/01/2017 I Started my internship at RL team at SAS Institute . SAS Institute
11/13/2016 I am attending INFORMS Annual Meeting 2016 conference at Nashville, TN!. I am presenting our Newsvendor paper. INFORMS16
08/04/2015 I serving as a system administrator in COR@L Lab, Lehigh ISE's high performance computing center. GlobalSIP
08/04/2015 Our paper, "Optimal scheduling of networked energy storages" accepted at IEEE Global Conference on Signal and Information Processing (GlobalSIP) conference at Orlando, FL!. GlobalSIP