Marco Cusumano-Towner

I am a third-year Ph.D. student in electrical engineering and computer science at MIT. I work with Vikash Mansinghka in the MIT Probabilistic Computing Project. I am supported by an NDSEG fellowship.

Previously, I worked at a clinical cancer diagnostics startup, where I developed computational infrastructure and applied machine learning algorithms for genetic testing based on DNA sequencing. During my Master's degree at Stanford, where I was funded by the NSF GRFP Fellowship, I worked on machine learning for predictive modeling of gene expression. During my undergraduate studies in at UC Berkeley, I worked with Pieter Abbeel on probabilistic and optimization techniques for household robotics.

marcoct [at] mit (dot) edu  /  CV  /  GitHub  /  Google Scholar

Research

I am interested in developing programming languages, software systems, user interfaces, algorithms, and theory that make it easier to construct, reason about, and use complex probabilistic computations. I am also interested in probabilistic artificial intelligence and theories of cognition based on probabilistic reasoning.

Gen: probabilistic programming with fast custom inference via code generation
Marco Cusumano-Towner, Vikash Mansinghka
Accepted to 2018 workshop on Machine Learning and Programming Languages (MAPL, co-located with PLDI).
Incremental inference for probabilistic programs
Marco Cusumano-Towner, Benjamin Bichsel, Timon Gehr, Martin Vechev, Vikash Mansinghka
Accepted to the 2018 ACM SIGPLAN Conference on Programming Language Design and Implementation (PLDI).
Using probabilistic programs as proposals
Marco Cusumano-Towner, Vikash Mansinghka
Presented at the 2018 Probabilistic Programming Semantics Workshop, co-located with POPL.
AIDE: An algorithm for measuring the accuracy of probabilistic inference algorithms
Marco Cusumano-Towner, Vikash Mansinghka
To appear in Advances in Neural Information Processing Systems 30, 2017.
Probabilistic programs for inferring the goals of autonomous agents
Marco Cusumano-Towner, Alexey Radul, David Wingate, Vikash Mansinghka
arXiv Preprint, 2017.
Encapsulating models and approximate inference programs in probabilistic modules
Marco Cusumano-Towner, Vikash Mansinghka
Presented at the 2017 Probabilistic Programming Semantics Workshop, co-located with POPL.
Measuring the non-asymptotic convergence of sequential Monte Carlo samplers using probabilistic programming
Marco Cusumano-Towner, Vikash Mansinghka
Presented at the 2016 NIPS Workshop on Advances in Approximate Bayesian Inference.
Quantifying the probable approximation error of probabilistic inference programs
Marco Cusumano-Towner, Vikash Mansinghka
arXiv Preprint, 2016.
A social network of hospital acquired infection built from electronic medical record data
Marco Cusumano-Towner, Daniel Li, Shanshan Tuo, Gomathi Krishnan, David Maslove
Journal of the American Medical Informatics Association 20.3 (2013): 427-434.
Bringing clothing into desired configurations with limited perception
Marco Cusumano-Towner, Arjun Singh, Stephen Miller, James O'Brien, Pieter Abbeel
Robotics and Automation (ICRA), 2011 IEEE International Conference on. IEEE, 2011.
Cloth grasp point detection based on multiple-view geometric cues with application to robotic towel folding
Jeremy Maitin-Shepard, Marco Cusumano-Towner, Jinna Lei, Pieter Abbeel
Robotics and Automation (ICRA), 2010 IEEE International Conference on. IEEE, 2010.

website template credit