I am a PhD candidate in Computer Science at Boston University pursuing research in machine learning and data mining, advised by Dr. Evimaria Terzi.
Before starting my PhD I worked for 3 years as a Data Scientist at IBM, where I developed machine learning solutions from the ground-up on exciting projects at the intersection of business and tech. I developed a bid-price optimization engine in Python, where I used hierarchical clustering and logistic model trees to build a microservice to handle real-time pricing requests. I also helped automate pre-surgical authorization for patients using NLP on clinical data. Prior to IBM I was at the University of Rochester, where I studied Physics and Electrical & Computer Engineering as an undergraduate, and explored computer audition (with Prof. Zhiyao Duan), computational physics and game theory. I then pursued a masters degree in Data Science with a focus in computational and statistical methods, and worked with Prof. Jiebo Luo.
My research interests are shaped by the diversity of my past experiences: I am excited about all things statistical, algorithmic and game theoretic. I am interested in studying about optimality and efficient strategies in environments of incomplete and imperfect information. I am curious about using machine learning and data-driven methods to formulate a better understanding of social systems. I am also interested in using data distributions to explain real-world phenomenon. Recently I became fascinated about sampling trajectories over expectation in stochastic games, and took an active interest in Reinforcement Learning: I built a Settlers of Catan boardgame framework with an AI player in Python.
I love to play (and watch) tennis and soccer. In my free time I challenge myself with chess, poker and other strategy games such as Catan and Avalon. I cook and tend to my growing plant collection to relax and express my artistic side. I enjoy hiking, biking, skiing and spending time outdoors.
Settlers of Catan boardgame built in Python. The goal of this project is to implement full multiplayer game functionality and use MCTS and reinforcement learning to build an AI player that can effectively explore-exploit heuristic strategies.
Xenou, Konstantia, Georgios Chalkiadakis, and Stergos Afantenos. "Deep Reinforcement Learning in Strategic Board Game Environments." European Conference on Multi-Agent Systems. Springer, Cham, 2018.
Gendre, Quentin, and Tomoyuki Kaneko. "Playing Catan with Cross-Dimensional Neural Network." International Conference on Neural Information Processing. Springer, Cham, 2020.
I collaborated with Prof. Jiebo Luo to research the spread of COVID-19 along political lines in US States using transportation patterns and weather data. I wrote code to cluster and analyze population mobility features across Red and Blue states (as per the 2016 presidential election) and discovered significant correlation patterns between the way people travel and the COVID-19 infection growth rate from March to July 2020.Code
I was awarded a research fellowship by Xerox to conduct research under the guidance of Prof. Zhiyao Duan in the Audio Information Research Lab at the University of Rochester during the summer of 2016. The primary purpose of my research was to design a computational system that could follow live musical performances and display pre-encoded lyrics in real-time. I implemented a multi-threaded audio framework with a shared synchronous buffer to handle simultaneous recording and processing. I extracted chroma feature vectors from the waveform audio data and then used online dynamic time warping to align the live audio stream with a pre-annotated version of the audio based on their harmonic similarity, and display lyrics in real-time.Code
Under the guidance of Prof. Jiebo Luo, I used efficient feature extraction on historical tennis match data, combined with a machine learning implementation in Python to predict the likelihood of professional tennis player success with 80% accuracy. I implemented code to create player-specific statistical feature sets aggregated from individual match data, and used neural network and logistic regression classification models to categorize player success.Code
An Ultimate Tic Tac Toe framework in Java, with an implementation of adversarial search using MiniMax with Alpha-Beta pruning. Ultimate Tic Tac Toe comprises nine 3x3 Tic Tac Toe boards, and the goal is to win 3 boards. I also developed a heuristic AI player, which was tested to beat a control player in 99 out of 100 games.Code
A complete implementation of the Enigma Machine in Python. I use an object-oriented framework to design the plugboard, reflector and rotor set, allowing for full encryption and decryption functionality. I also implement code to crack the Enigma cipher, using a known-plaintext attack methodology.Code
I investigated the non-linear dynamics of the damped and driven pendulum and developed a theoretical framework for the system. I then computationally solved the classical mechanics problem using Mathematica to discover regions of chaotic and non-chaotic motion.Paper
Designed, built and tested binaural headphones with real-time recording and filtering capability, with a 12ms latency. Modeled the head-related tranfer function (HRTF) using a Neumann head microphone, and implemented the FFT algorithm in C++ to enable real-time audio filtering.Presentation
A statistical framework in Python to predict stock price evolution using geometric Brownian motion. The model was tested to have under 5% error using Monte Carlo simulations on 2 years of historical Nike stock prices.
Reference: Dmouj, Abdelmoula. "Stock price modelling: Theory and Practice." Masters Degree Thesis, Vrije Universiteit (2006).
Computer Science PhD candidate pursuing research in machine learning and data mining.
At IBM, I have designed real-world solutions and developed ML models from the ground up on exciting projects at the intersection of business and tech. Under the guidance of Dr. Mark Freeman, I developed a novel algorithmic method in Python for business-to-business
bid price optimization, and successfully deployed it as a microservice for a large telecommunications company.
I have also worked with Watson Health, to develop an NLP solution to automate pre-surgical authorization. My contributions included building a rules engine in Python and implementing algorithms to extract natural language features from patient clinical data.
I worked with 2017 Facebook ad-campaign data to help predict drivers of campaign success and identify optimal ad-campaign configurations, using SQL scripts to generate campaign performance distributions. I then developed classification models in Python to predict campaign-optimizing key performance indicators and presented a metric-driven, ad-campaign configuration process to the company.
I facilitated workshops, recitations and labs for groups of 15-20 students for 9 different courses across the Physics, Electrical & Computer Engineering and Math departments. I was awarded the Citation for Achievement in College Leadership in recognition of outstanding undergraduate teaching and research commitment.
Under the supervision of Dr. Ravi Venkatramani, I investigated conformation changes and molecular dynamics in the photoisomerization cycle of Bacteriorhodopsin using NAMD simulation software.
Focus in Machine Learning and Data Mining
Focus in Computational & Statistical Methods
Relevant coursework: Artificial Intelligence, Statistical Learning, Data Mining, Database Systems, Cryptography
Highest Distinction, Magna Cum Laude
Relevant coursework: Computational Physics, Data Structures & Algorithms, Robot Control, Game Theory, Classical Mechanics, Linear Algebra, Signal Processing