B.A. in Computer Science and Mathematics from Hamilton College
Machine Learning Engineer, Entrepreneur, Lover of the Outdoors
I'm a machine learning research engineer with an entrepreneurial spirit. I enjoy pool, chess, and exploring the great outdoors - from hiking to skiing. I have the most language experience working with Python and C++, and have used many others - Java, Swift, JavaScript, etc. - on various personal, academic, and professional projects. I have recent experience with designing and developing novel methods in AI and ML. I also have experience applying state-of-the-art methods in ML to solve novel problems in industry in intuitive ways. My passion lies in developing a deeper theoretical understanding of machine learning using data from the brain. I always jump at the opportunity to learn new skills and develop new technologies, whatever that might entail.
The low availability of high quality, labeled datasets of panchromatic satellite imagery featuring passenger vehicles for both academic and commercial applications led to the creation of CiPSI, a new dataset to fill exactly that gap. Using this new dataset, we evaluated six existing state-of-the-art classification networks and three detection methods. Ultimately, we introduce a novel network architecture, CiPSINet, that achieves higher performance than all other methods for vehicle classification while using fewer learnable parameters. Further, using this network, we improve the detection accuracy of the Faster R-CNN method by replacing the ResNet-50 backbone with CiPSINet.
Starting as research that earned me honors for my major in computer science, Phased Local Iterative Search for k-plex Enumeration (PLISkE) is a novel search algorithm for enumeration of dense subgraphs. PLISkE has proven applications in Video Object Co-Segmentation (20% improvement over existing framework), Feature Filtering in Machine Learning datasets (improvements in speed of search and quality of results over existing), and a novel application in the geosciences that cuts hours of manual work in HALite (below) down to seconds of waiting time.