M.S. Computer Science

I completed a M.S. in Computer Science from Georgia Institute of Technology, specializing in Robotics & Perception and Machine Learning in the OMSCS program (Online Master of Science in Computer Science).

Courses

  • Artificial Intelligence
  • Artificial Intelligence for Robotics
  • Computability, Complexity, & Algorithms
  • Computational Photography
  • Computer Vision
  • Human-Computer Interaction
  • Knowledge-Based AI
  • Machine Learning
  • Machine Learning for Trading
  • Reinforcement Learning

Artificial Intelligence

Adversarial AI Game Playing

Using MINIMAX, alpha-beta pruning, and heuristic evaluation functions, I created an AI agent that can play the 2-player game "Isolation" with 4 queens better than a human.

Raven's Progressive Matrices

I constructed an AI agent to address a human intelligence test. The agent was tasked with interpreting visual representations, deciphering patterns, and decision-making.

SAMPLE IMAGES

Raven's Progressive Matrices

The agent is presented with figures A-H and must select the appropriate figure from 1-8.

The agent is presented with figures A-H and must select the appropriate figure from 1-8.

The agent is presented with figures A-H and must select the appropriate figure from 1-8.

The agent calculated the similarity of each unary relationship and applied its understanding to determine the pattern.

Constraint Satisfaction Problems

For solving constraint satisfaction problems on finite domains, I learned techniques including backtracking, constraint propagation, and local search.

Hidden Markov Models

This project implemented the Viterbi and Forward-Backward algorithm to recognize morse code signals using HMMs.

Machine Learning

Decision Trees & Forests

I completed a project to build, train, and test several decision tree models to perform basic classification tasks.

Supervised Learning Techniques

I compared five supervised learning techniques: decision trees, boosting, k-nearest neighbors, support vector machines, and neural networks.

GitHub

Probability & Bayesian Networks

To gain a better understanding of probabilistic systems, I implemented Bayesian networks and Gibbs & Metropolis-Hastings sampling algorithms.

Unsupervised Learning

This code explores 2 clustering algorithms (k-means & expectation-maximization) and 4 dimensionality reduction algorithms (PCA, ICA, Randomized Projections, & Factor Analysis).

GitHub

Machine Learning for Trading

With a focus is on how to apply probabilistic machine learning approaches to trading decisions, I applied various algorithms like linear regression and Q-Learning to stock trading situations.

Markov Decision Processes

This repository explores two Markov Decision Processes (MDPs) and implements three algorithms: value iteration, policy iteration, and Q-learning. The problems and algorithms are compared in terms of convergence, iterations, runtime, and optimal rewards.

GitHub

Computer Vision

This course developed basic methods for computer vision including fundamentals of image formation, camera imaging geometry, feature detection and matching, stereo, camera calibration, motion estimation and tracking, image classification and scene understanding.

Projects

Human-Computer Interaction

The course covered three broad categories of topics within human-computer interaction:

  • the principles and characteristics of the interaction between humans and computers
  • the techniques for designing and evaluating user-centered systems
  • current areas of cutting-edge research and development in human-computer interaction

Computability, Complexity, & Algorithms

This course covered concepts from computability theory:

  • Decidability, Countability, and Reducibility
  • NP-Completeness
  • Time & Space Complexity
  • Combinatorial & Algebraic Algorithms
  • Max-flow / Min-cut and Bipartite Matching