Justin Solomon, an associate professor in the Department of Electrical Engineering and Computer Science at the Massachusetts Institute of Technology (MIT), is making use of modern geometric techniques to address intricate problems, many of which don’t appear to be linked to shapes. He extrapolates from the foundations laid more than 2,000 years ago by the Greek mathematician Euclid to apply geometric data processing to an array of datasets that share some geometric structure.
For instance, a statistician may want to compare datasets to evaluate the impact on the performance of a machine-learning model. The examination of these datasets using geometrical tools can reveal whether the same model will function on both sets.
Two broader areas make use of Solomon’s geometry techniques. Around half of his team processes two- and three-dimensional geometric data, such as aligning 3D organ scans in medical imaging or helping autonomous vehicles identify pedestrians using spatial data from LiDAR sensors. The other half applies geometric tools for high-dimensional statistical research, including the construction of superior generative AI models.
Growing up in Virginia, Solomon started early with a keen interest in computer graphics. This lead him to intern at a research lab working on algorithms for 3D face recognition, and double-major in math and computer science at Stanford University. It was at Stanford, while developing algorithms for visual effects at Pixar every summer, he decided to launch an academic career, culminating in a PhD in computer science.
Solomon’s study centered on optimal transport, which looks for the most efficient way to move a distribution. His research on graphics applications extended to other areas and applications, many surprisingly unrelated. This unexpected broadening inspired the configuration of his research group at MIT.
Keen to engage with students and peers at MIT, Solomon hoped to offer practical solutions to complex problems affecting various disciplines. He set out to make geometric research accessible to those who are often underserved by it, launching the Summer Geometry Initiative, a six-week paid research program focusing on undergraduates from underrepresented backgrounds. The program, now in its third year, has begun to see tangible results, including the increased diversification of incoming classes of PhD students at MIT and other institutions.
In the ever-growing field of problems within machine learning and statistics that can be resolved through geometric techniques, Solomon emphasizes the need for a diverse range of researchers. He is intrigued by and eager to continue using geometric tools to enhance unsupervised machine learning models, where models learn to recognize patterns without labeled training data.
When he’s not deeply engaged in complicated research problems, Solomon enjoys classical music and plays the cello with the New Philharmonia Orchestra in Newton, Massachusetts. This connection between the analytical nature of music and the context of his computer graphics research provides a unique harmony to his interests.
