cv

Bachelor of Science

Howard University

Mathematics, Concentration in Applied Mathematics

• Minor: Computer Science

Undergraduate Researcher (Geometric Optics)

Howard University

Supervisor: Dr. Henok Mawi

• Investigating the Prescribed Intensity Problem, a non-linear inverse problem focused on designing freeform lens geometries that map source light distributions to target measures.
• Bridging classical optics with Optimal Transport theory, utilizing the Monge-Ampère Equation to model the transformation of probability densities via refraction and reflection.
• Developed a computational framework in MATLAB for simulating ray-tracing on non-standard surfaces; currently transitioning to numerical solvers for non-linear PDEs.

Vote Estimator (Ecological Inference)

Developing a custom Ecological Inference (EI) model to estimate latent vote probability vectors from aggregate precinct data.

• Formulated a compound loss function combining L2 data-fit, neighbor-weighted Laplacian smoothing, and entropy regularization to solve for non-degenerate probability distributions.
• Implementing convex optimization via the Adam optimizer to minimize loss over the probability simplex, utilizing PyQGIS to construct spatial adjacency matrices.

SpadesAI (Reinforcement Learning)

Engineered a game-playing agent using Monte Carlo Tree Search (MCTS) guided by deep neural networks, similar to the AlphaZero architecture.

• Implemented PUCT (Predictor Upper Confidence Trees) to balance exploration/exploitation across a state space of approximately 10^20.
• Trained three distinct neural networks (Bidding, Playing, and Value) via self-play reinforcement learning, achieving a 2100+ Elo rating against human players.

Youth Coordinator

East Side Parkways Coalition

• Organized community canvassing efforts to support the restoration of Humboldt Parkway.
• Assisted in fundraising efforts supporting legal action against the NYSDOT.
• Contributed to a campaign that successfully resulted in a court injunction requiring an Environmental Impact Statement (EIS).