Over 2,000 years ago, Greek mathematician Euclid, often called the father of geometry, revolutionized the understanding of shapes. In today’s technological era, a 21st-century geometer, Justin Solomon, uses sophisticated techniques to solve complex problems related to shapes but often unrelated to them. Solomon applies geometry to study datasets for comparing their effectiveness in machine learning models by probing their space, structure, and arrangement.
Justin Solomon is an associate professor in the MIT Department of Electrical Engineering and Computer Science (EECS) and a member of the Computer Science and Artificial Intelligence Laboratory (CSAIL). Solomon’s work focuses on applications of geometric tools to understand relationships, similarities, curvature, and shape in datasets. The idea is that data often has geometric properties due to the arrangement in high-dimensional space. The discoveries Solomon and his team could make would help determine the model’s effectiveness on the datasets.
Solomon’s Geometric Data Processing Group is divided into two main research areas. Half the team focuses on problems involving processing 2D and 3D geometric data, such as aligning 3D organ scans in medical imaging or helping autonomous vehicles identify pedestrians from spatial data collected by LiDAR sensors. The other half use geometric tools to study high-dimensional statistics, such as improving generative AI models.
Born and raised in Northern Virginia, Solomon had an early interest in computer graphics, which led him to intern at a research lab outside Washington, where he helped develop algorithms for 3D face recognition. He went on to study math and computer science at Stanford University, where he interned at Pixar Animation Studios. At Pixar, he worked on enhancing realism in animated films and transforming the “look” of animated content.
Later, Solomon earned his PhD in computer science from Stanford, focusing on the optimal transport problem involving moving a distribution efficiently from one location to another. His research expanded beyond computer graphics applications of optimal transport to other areas, ultimately shaping the structure of his research group at MIT.
At MIT, Solomon found the perfect environment to work on intricate and practical problems alongside bright students, postdocs, and colleagues. He also realized the need to make geometric research more accessible to people from diverse backgrounds. Solomon initiated the Summer Geometry Initiative, a six-week paid research program for undergraduates from underrepresented backgrounds. This program, now in its third year, has seen significant results in change the composition of incoming PhD classes, not only at MIT but also in other institutions.
Solomon sees potential to address an increasing number of problems in machine learning and statistics using geometric techniques. This intersection of geometric principles and machine learning could enhance unsupervised machine learning models and help them recognize patterns without labeled training data.
Outside of research, Solomon is an enthusiastic musician, playing classical music on piano and cello. He plays cello with the New Philharmonia Orchestra in Newton, Massachusetts. According to Solomon, music’s analytical nature beautifully complements his work in computer graphics.
