I am currently pursuing a masters degree in Computer Science at Georgia Institute of Technology (Georgia Tech) with a specialization in Machine Learning. To learn more about my background, see About Me.
The following projects were completed by me, independently. Many were completed for my master program at Georgia Tech.
| Name | Topics | Description | Language |
|---|---|---|---|
| Machine Learning (Supervised) | Scikit-learn, Supervised Learning | Trained neural networks (MLPClassifier), decision trees (DecisionTreeClassifier), KNN classifiers (KNeighborsClassifier), and ensemble learners (AdaBoostClassifier) to compare and contrast model performance across two datasets: Rain in Australia and Bank Marketing. | Python |
| Machine Learning (Unsupervised) | K-means clustering, Expected Maximization, PCA, ICA, Randomized Projection, Feature Importance, BIC plot, TSNE plots | Used a variety of unsupervised learning techniques to analyze two datasets ( Rain in Australia and Bank Marketing) and perform dimensionality reduction. | Python |
| Optimization | Randomized Optimization, Gradient Descent, Reinforcement Learning | Compared and contrasted gradient descent, randomized hill climbing, simulated annealing, a genetic algorithm, and MIMIC performance when solving the following problems discrete optimization problems: One Max, NQueens, and FlipFlop. | Python |
| Augmented Reality for Ad Projection | Computer Vision, Augmented Reality, Feature Detection, Distance Transform, OpenCV | Created a program that can project an advertisement onto a wall in real time (30 frames per second) to create an augmented reality experience. The ad remains stable when the camera is moved. Compared visual performance and processing times when using AKAZE and ORB feature detection methods. | Python |
| Assessing Learners | Decision Trees, Random Trees, and Random Forests | Comparing in and out of sample error and training and query times to assess tree machine learning architectures, including decision trees, random trees, and random forests. | Python |
| Robot Localization and Navigation | Kalman Filters, Particles Filters (Sequential Monte Carlo (SMC) methods), A* search algorithm | Implemented filters (including Kalman and particle filters) for robot localization and search algorithms (including A*) to plan the shortest path from one point to another. | Python |
| Reinforcement Learning with Q Learner | Reinforcement Learning, Q Learner, Dyna-Q, Robot Navigation | Implemented Q-Learning and Dyna-Q solutions for a reinforcement learning problem - robot navigation. | Python |
| Distributed File System | GRPC, sockets, multi-threaded programming | Created a distributed file system using protocol buffers and gRPC. Incorporate a weakly consistent synchronization system to manage cache consistency between multiple clients and a single server | C++ |
| Using semantic networks to solve human intelligence tests | Semantic networks, artificial intelligence | Solved tests of human intelligence (modified Raven Progressive Matrices tests) using semantic networks | Python |
| Word Smith | Object Oriented Programming, Trie Data Structure, Data Structures and Algorithms | A word-construction, multi-player game using a prefix trie data structure. | C++ |