Matt O'Shaughnessy
School of Electrical & Computer Engineering
Georgia Institute of Technology
Email: moshaughnessy6@gatech.edu
About me
I recently completed my Ph.D. in the Center for Machine Learning at Georgia Tech, where I was very fortunate to be advised by Mark Davenport and Chris Rozell, and to be supported by the NDSEG Fellowship.My research develops mathematical tools that use structure hidden in data to help scientists understand complex systems. More specifically, my work involves causality and low-dimensional structure; current and past projects involve (1) theory and algorithms for causal inference in dynamical systems; (2) causality-inspired methods for explaining black-box classifiers; and (3) Bayesian methods for exploiting temporal and low-dimensional structure in inverse problems.
I'm interested in science and technology policy, and am particularly interested in the impacts of artificial intelligence on inequality and democracy. In addition to my technical work, my research has explored public opinion on AI adoption and governance, societal impacts of machine-learning-enabled disinformation, and governance strategies for neurotechnologies. I'm also involved in IEEE-USA's AI policy committee, where I co-chair the subcommittee on Democratic Use of AI.
Outside of research, I enjoy hiking and cycling, music, and baking. I have an only slightly tenuous claim to an Erdös-Bacon-Sabbath number of eight: my Erdös number is 3, my Bacon number (from a role as an extra in a major movie) is 2, and my Sabbath number (as a former violist) is 3.
Selected papers (machine learning)
- M. O'Shaughnessy, M. Davenport, and C. Rozell, "Distance preservation in state-space methods for detecting causal interactions in dynamical systems," Submitted.
- M. O'Shaughnessy, G. Canal, M. Connor, M. Davenport, and C. Rozell, "Generative causal explanations of black-box classifiers," Proc. Advances in Neural Information Processing Systems (NeurIPS), December 2020.
- M. O'Shaughnessy, M. Davenport, and C. Rozell, "Sparse Bayesian learning with dynamic filtering for inference of time-varying sparse signals," IEEE Transactions on Signal Processing, December 2019.
Selected papers (science & technology policy)
- M. O'Shaughnessy, W. Johnson, L. Tournas, C. Rozell, and K. Rommelfanger, "Neuroethics guidance documents: Principles, indicators, and implementation strategies," Submitted.
- M. O'Shaughnessy, D. Schiff, L. Varshney, C. Rozell, and M. Davenport, "What governs public opinion on AI adoption and governance?" Submitted.
- M. O'Shaughnessy, "Security implications of machine learning enabled disinformation," 2020.