Matthew Graham

California Institute of Technology

Research Professor of Astronomy

I am interested in the application of machine learning and other advanced statistical techniques to astrophysical time series, particularly the study of stochastic/aperiodic variability in astronomical populations, e.g., AGN. In recent years, I have employed RNNs, Bayesian blocks, and Slepian wavelets, for example, to study extreme AGN optical variability and SMBH binaries, significant flaring events, changing-look quasars, and candidate EM counterparts to compact object mergers in AGN disks. I am the Project Scientist for the Zwicky Transient Facility (with several hundred thousand alerts per night already) and have more general interests in both transient and variable phenomena and associated infrastructure and am the PI for one of the approved LSST Alert Brokers (Babamul). I am also involved with the NSF HDR Institute for Accelerated AI Algorithms for Data-Driven Discovery which looks at both hardware and software solutions to low latency inferencing. I am a member of both the ISSC and AGN SC. 

Contact Matthew Graham