About me
I am Reza Alvandi, a master’s student in electrical and computer engineering.
My interests lie at the intersection of Machine Learning, Statistics, and Optimization. Having previously studied computer science, I am also passionate about algorithms and software development. I focus on designing intelligent systems with mathematical guarantees, particularly for control tasks, using tools such as game theory and machine learning.
Research Interests
- Decentralized stochastic control and Markov decision theory
- Reinforcement learning for partially observed systems
- Theoretical Machine learning
Education
- McGill University, Montreal, Canada
- Degree: Master of Science
- Department: Computer and Electrical Engineering, Electrical Engineering
- Duration: Jan. 2025 - Dec. 2026 (Expected)
- Sharif University of Technology, Tehran, Iran
- Degree: Bachelor of Science
- Department: Mathematical Sciences, Computer Science
- Duration: Sep. 2019 – Feb. 2024
My Projects
Active Contours for Object Detection
Active Contours (Snakes) for object detection. Define an energy functional comprising internal and external energies, iteratively minimize the energy function using Dynamic Programming to deform the contour towards the object boundaries.
.gif)
TweetGram Social Media in Java
Our Image Morphing project, combines linear algebra concepts and the Delaunay Triangulation algorithm for creating stunning image transitions. Delaunay Triangulation is a fascinating and efficient technique used to transform a set of points into a non-overlapping set of triangles.
Adversarial Search Board Game Player
Using Minimax with Alpha-Beta Pruning and Heuristic Evaluation Functions to design an agent that can win board games.
Autonomous Precision: Deep Q Learning Mastery in Spaceship Simulator Landings
I designed a Deep Q Learning algorithm that adeptly lands a spaceship in a simulator, showcasing efficient autonomous navigation and control.
Image Blending (Gradient Domain Blending)
Using a carefully designed mask to extract pixel values from the first image and image derivatives from the second image. Utilized sparse methods to efficiently solve the resulting large-scale linear system, resulting in visually captivating compositions.
Vehicle Insurance Simulation
Created a simulation model for a vehicle insurance organization, optimizing staffing and queuing to enhance the process efficiency. Analyzed event types, service time distributions, and queuing discipline to inform decision-making.
Cluster System Computation Server in Java
Developed a cluster system to distribute computations across multiple worker servers. Utilized caching for efficient task handling, parallel processing, and load distribution through a centralized server, showcasing expertise in distributed computing and optimization.