Chicago, Illinois, United States
• Applied Mathematician specializing in applied harmonic analysis, differential geometry, and the statistical theory of machine learning. • Computer Scientist with broad knowledge in algorithm design, geometry processing, and data visualization. • Instructor of two undergraduate multivariable calculus courses and a research independent study.
• Prototyped in Python a high-accuracy real-time machine learning system for malicious URL detection based on Random Forest. Implementation involved extensive programming with Spark Streaming and MLlib; final product deployed to Amazon EC2. • Built a Titan graph database from real web crawler data, and created a Python demo for infringement prediction based on graph inference algorithm Loopy Belief Propagation. The database is highly availalable through Rexster Graph Server, backed by Cassandra and ElasticSearch.
My Ph.D. research focused on understanding the geometry of massive high-dimensional datasets. In particular, I extended the state-of-the-art dimensionality reduction technique to the next level, and built up a novel theoretical framework for semi-supervised manifold learning. My M.S. in Computer Science involved applications of this general framework for real data problems arsing from geometric morphometics (evolutionary anthropology based on morphological traits, with modern techniques), plus extensive coursework. • Applied Mathematics and Machine Learning: -- Diffusion Geometry -- Differential Geometry -- Metric Space Approximation -- High-performance Scientific Computing • Computer Science: -- Computer Graphics -- Geometry Processing -- Data Visualization -- Probabilistic Graphical Models • Technical Skills: -- Python/MATLAB/C++ programming in Linux/Unix environment; -- Scala/Java/Processing/Javascript programming and web design with HTML/CSS; -- Big Data frameworks such as Spark (MLlib, GraphX, Streaming) and Hadoop (MapReduce)