Computer-Aided Discovery in Astronomy
Next-generation astronomy will face rapidly growing data volumes from ground-based and space-based telescope networks. In radio astronomy for instance, high data rates from antenna arrays such the SKA will pose new challenges to data analysis. As real-world phenomena are digitized and mapped to data, the scientific discovery process essentially becomes a search process across multidimensional Big Data sets. The extraction of meaningful discoveries from this sea of data therefore requires highly efficient and scalable machine assistance to enhance human contextual understanding.
This talk will outline new directions for Computer-Aided Discovery that aim to address these challenges. Scientists will depend on enhanced automation, especially for tasks that require matching of a multitude of theoretical models with concrete empirical observations. The ongoing work at MIT Haystack supported by NSF and NASA opens up new possibilities for a computational infrastructure to answer questions such as: What inferences can be drawn from an identified feature? What does a finding mean and how does it fit into a big theoretical picture? Does an observation contradict or confirm previously established models and findings? How to test hypotheses and ideas effectively? The presentation will elaborate on conceptual and technical approaches discuss perspectives for the future.
————-
Dr. Victor Pankratius
Dr. Pankratius leads the Astro- & Geoinformatics group at the Massachusetts Institute of Technology, Haystack Observatory. He is a computer scientist dedicated to interdisciplinary research, and he currently serves as a principal investigator in several interdisciplinary projects funded by NSF and NASA. He is also a collaborative partner of MIT’s Computer Science and Artificial Intelligence Laboratory (CSAIL) and the Department of Earth, Atmospheric, and Planetary Sciences (EAPS).
Dr. Pankratius’ computer science interests include multicore parallel systems and software engineering. Past contributions covered automation in performance tuning, parallel program debugging, and empirical studies in programming language productivity. Collaborations with industry partners such as Intel, Sun Labs, Oracle helped him develop a better understanding of complex parallel systems in practice. In astronomy, Dr. Pankratius is involved in software aspects of the ALMA Phasing project to enhance the ALMA Observatory with Very-Long Baseline Interferometry capabilities, software imaging on Event Horizon Telescope, and the Radio Array of Portable Interferometric Detectors (RAPID).
In the scientific community, Dr. Pankratius chaired a variety of workshops and conferences, such as the ACM ICSE workshop series on Multicore Software Engineering and the International Conference on Multicore Software Engineering, Performance, and Tools, which emerged from these series. He also co-edited special issues of IEEE Software and IEEE Computer on parallel computing and computing in astronomy, respectively.
Dr. Pankratius holds a Habilitation degree in Computer Science from the Karlsruhe Institute of Technology and a Dr.rer.pol. degree with distinction from the University of Karlsruhe. From the University of Münster, Germany, he received a Diplom degree (M.S.) in Business Computer Science best of class and a Bachelor of Science in Information Systems (BScIS).
Contact Dr. Pankratius via Email at pankrat [at] mit.edu, on the Web at www.victorpankratius.com, or on Twitter @vpankratius.
Open Parallel 