Year 1 (2019-2020)

Major Activities

Research collaboration; Research training at all levels; Establishing partnerships with ongoing local data science and machine learning organizations and activities, both academic and in industry.

Significant Results

  • Machine Learning in Medicine Seminars (virtual) has been expanded to include our institute and both universities.
  • Machine Learning in Medicine Symposium planning is underway but will need to move to a virtual format.
  • Partnerships with the University of Rochester Goergen Institute for Data Science (GIDS) and the Cornell Center for Data Science for Enterprise and Society have been established.
  • REU for mathematics of data science is being run (virtually) now. Click here for details.

Key Outcomes or Other Achievements

  • Establishment of a COVID-19 working group, published paper and had significant local media coverage of this research effort.
  • Tripods NSF REU/Grad for All 2020: The goal of Grad For All is to empower and further inspire all interested students from colleges and universities in Western New York area to pursue advanced degrees in all fields. We are especially looking to involve traditionally under-represented groups, including women and underrepresented minorities. Our goal is to provide these students with the information, skills and training needed to succeed in graduate school, academic careers, and industry. Click here for details.
  • CAMSAP’19 tutorial: Connecting The Dots: Identifying Network Structure Of Complex Data Via Graph Signal Processing https://camsap19.ig.umons.ac.be/tutorials.php

Selected Publications

  • Wagner, A. B., Hill, E. L., Ryan, S. E., Sun, Z., Deng, G., Bhadane, S., Martinez, V. H., Wu, P., Li, D., Anand, A., Acharya, J., & Matteson, D. S. (2020). Social distancing merely stabilized COVID-19 in the US. Stat (International Statistical Institute), e302. Advance online publication. https://doi.org/10.1002/sta4.302
  • Blanca, A., Chen, Z., Stefankovic, D., and Vigoda, E. (2020). Hardness of Identity Testing for Restricted Boltzmann Machines and Potts models. Proceedings of Machine Learning Research, vol 125, pages 514-529. PMLR.
  • Davidow, M. and Matteson, D. S. (2020). Factor analysis of mixed data for anomaly detection. arXiv preprint arXiv:2005.12129.
  • Frank, A. S., Lupattelli, A., Matteson, D. S., Meltzer, H. M., and Nordeng, H. (2020a). Thyroid hormone replacement therapy patterns in pregnant women and perinatal outcomes in the o spring. Pharmacoepidemiology and Drug Safety, 29(1):111-121.
  • Frank, A.-S. J., Matteson, D. S., Solvang, H. K., Lupattelli, A., and Nordeng, H. (2020b). Extending balance assessment for the generalized propensity score under multiple imputation. Epidemiologic Methods, 9(1).
  • Gelsinger, M. L., Tupper, L. L., and Matteson, D. S. (2019). Cell line classification using electric cell-substrate impedance sensing (ecis). The International Journal of Biostatistics, 16(1).
  • Tang, B. and Matteson, D. S. (2020). Graph-based continual learning. arXiv preprint arXiv:2007.04813.
  • Wu, H. and Matteson, D. S. (2020). Adaptive bayesian changepoint analysis and local anomaly detection.
  • Zhang, W., Grin, M., and Matteson, D. S. (2020). Modeling nonlinear growth followed by long-memory equilibrium with unknown change point. arXiv preprint arXiv:2007.09417.