About Me

I'm currently a Masters student in Computing Science at University of Alberta focusing on neurosymbolic programming and reinforcement learning working with Levi Lelis and Sandra Zilles. Before that, I completed my B.Sc. in Computer Engineering at Amirkabir University of Technology.

Education

M.Sc. Computing Science

University of Alberta, Edmonton, AB

2023 – Expected 2025 | Supervisor: Levi Lelis and Sandra Zilles | GPA: 3.8

B.Sc. Computer Engineering (AI)

Amirkabir University of Technology, Tehran, Iran

2018 – 2023 | Major: AI; Minor: Networks | GPA: 18.75/20 (3.96/4)

Publication

• Common Benchmarks Undervalue the Generalization Power of Programmatic Policies, NeurIPS Track Position (under review), 2025

  Amirhossein Rajabpour, Kiarash Aghakasiri, Sandra Zilles, Levi Lelis.

  PDF Code Page

Research Experience

• Research Assistant | University of Alberta (May 2023 – Present)

  Conducted research on the generalization capabilities of neural versus programmatic policies that challenges the prevailing view that programmatic policies generalize better on out-of-distribution (OOD) tasks. Revisiting four influential benchmarks and showed that with small changes—such as simpler neural architectures, sparse observations, and safer reward functions—neural policies match or outperform programmatic ones in OOD generalization. Also proposed new benchmark ideas that highlight fundamental strengths of programmatic representations, such as memory-based reasoning. Supervised by Levi Lelis and Sandra Zilles. [Code]

• Individual Study Course Project | University of Alberta (Jan – Apr 2024)

  Worked on various ways—Implemented hill-climbing search, hill-climbing with exploration, and genetic algorithm—to improve the work "UNVEILING OPTIONS WITH NEURAL DECOMPOSITION". Supervised by Levi Lelis.

• RL1 Course Project | University of Alberta (Fall 2023)

  Comparing the performance of an expert agent, which uses all computational resources on a single task, against a generalist agent that distributes the same resources across multiple tasks. The study evaluates their performance across various reinforcement learning algorithms like PPO, DQN, and APPO, and explores the impacts of different environments, MiniGrid scenarios, and budgets (Project Report).

• Bachelor’s Thesis | Amirkabir University of Technology (Mar – May 2023)

  Designed a face aging platform using CycleGAN and reversible models. Supervised by Mohammad Rahmati. [Code]

• Research Assistant | Amirkabir University of Technology (Feb – Jun 2022)

  Implemented GCN for portfolio asset allocation on time-series data. Supervised by Hamed Farbeh.

Work Experience

• Technical Advisement | Alberta MAchine Intelligence Institute (AMII) (Sep – Oct 2024)

  Conducting research on advanced AI/ML methods to enhance real-time intrusion detection and anomaly detection in high-speed network storage infrastructures. Focused on improving threat detection accuracy and scalability in multi-cloud environments.

• Machine Learning Engineer | Crouse PJS Co. (Oct 2021 – Jan 2022)

  Working on an Artificial Intelligence Vision problem. My job was to design a light machine learning model to do the following to recognize whether LEDs on the monitor work fine:

  - Localizing LEDs on the monitor

  - Clustering LED light pixels and using the most valuable clusters for extracting light information with fuzzy C-means clustering

  - Extracting luminance and wavelength from those selected clusters

  - Applying different regression models for each LED color to detect malfunctioning LEDs

• DevOps Engineer Intern | Graph Co. (Nov 2020 – Apr 2021)

  Streamlined workflows using Docker, Minikube, and backend tooling.

Technical Skills

Teaching Assistant

Graduate Teaching Experience | University of Alberta

• CMPUT 366: Search & Planning in AI | Fall 2024, Winter 2025

  Lectures; labs; grading projects, assignments, exams & quizzes.

  Instructor: Levi Lelis

• CMPUT 291: Introduction to File & Database Management | Fall 2023, Winter 2024

  Lectures; labs; grading projects, assignments & exams.

  Instructors: Davood Rafiei & Arash Dargahi Nobari

Undergraduate Teaching Experience | Amirkabir University of Technology

• Principles of AI | Fall 2022

  Project design & grading on CSPs, adversarial search, Bayesian nets, RL.

  Instructor: Mahdi Javanmardi

• Cloud Computing | Fall 2022

  Design and grade assignments & labs on APIs, cloud services, Docker & Kubernetes.

  Instructor: S. Ahmad Javadi

• Internet of Things | Fall 2022

  Designing (and grading) assignments about various scenarios for working with different sensors and actuators and programming with Arduino.

  Instructor: Siavash Khorsandi

• Algorithm Design | Winter 2021, Fall 2021, Winter 2022

  Problem sets on greedy, divide & conquer, dynamic programming.

  Instructors: Alireza Bagheri & Sajad Shirali-Shahreza

Selected Graduate Courses

• Reinforcement Learning 1 (Marlos Machado)
• Reinforcement Learning 2 (Rich Sutton)
• Modeling Strategic Behavior (James Wright)
• Machine Learning (Lili Mou)

Honors & Awards

• Fully funded admission to UAlberta CS, ECE & Radiology; University of Western Ontario CS (2023)
• 4-year scholarship, Amirkabir University of Technology (2019–2023)
• Top 1% in Nationwide Math University Entrance Exam (2018)