More than two millennia ago, Greek mathematician Euclid, seen by many as the originator of geometry, significantly influenced our comprehension of shapes. Fast forward to today, Justin Solomon, using advanced geometric techniques, is resolving challenging issues that appear unrelated to shapes on the surface.
Solomon, an associate professor in the MIT Department of Electrical Engineering and Computer Science and member of the Computer Science and Artificial Intelligence Laboratory, is applying parallels in geometry to solve significant high-dimensional problems. A prime example could be a statistics professional wishing to compare two datasets to assess how using one for training and another for testing could influence the efficacy of a machine learning algorithm.
This comparison becomes a geometric challenge if the datasets’ contents share geometric alignment based on data arrangement in high-dimensional space. If successfully navigated using geometric tools, this could offer insights into whether the same model would work on both sets.
Approximately half of Solomon’s research team works on challenges involving two- and three-dimensional geometric data processing. This includes aligning 3D organ scans in medical imaging or facilitating autonomous vehicles to identify pedestrians using spatial data from LiDAR sensors.
Whereas, the remaining team members conduct high-dimensional statistical research using geometric tools to construct better generative AI models. In real terms, this might include developing algorithms for generating new images from a dataset filled with existing images which, is essentially, a geometric problem.
Solomon’s interest in computer graphics started at a young age. This led him to gain an internship in a Washington-based research lab during high school which only cemented his passion for the field. He pursued a double-major in math and computer science at Stanford University. Driven by both his enthusiasm for the field and the unique mathematical challenges it presents, Solomon proceeded to intern at Pixar Animation Studios throughout college and into graduate school.
Ultimately, this led him to focus his career within academia, specifically specializing in a problem known as optimal transport during his PhD studies. This area of study looks at distributing items as efficiently as possible, often looking at minimizing cost.
The success of his findings saw him drawn to MIT, where he had the opportunity to work on complex problems with fellow professionals and students alike. He has since made it his mission to broaden the reach of geometric research to underserved students. To achieve this, he launched the Summer Geometry Initiative, a six-week paid research program for undergraduates drawn predominantly from underrepresented backgrounds.
This desire to help the next generation of researchers reflects Solomon’s commitment not only to his students but also to his field. He is excited about improving unsupervised machine learning models by applying geometric tools, that can help in complex, unlabeled 3D scenes.
Outside of academia, Solomon is a keen musician and uses this to further his interests, actively participating in a local symphony in each city he resides in. He currently plays the cello with the New Philharmonia Orchestra in Newton, Massachusetts. To Solomon, the practice of music is intrinsically analytical in nature and closely connected to his work in computer graphics. Thus, he sees the two interests as mutually beneficial.
