Thursday November 5, 2009
1:00 - 2:00 pm (10:00-11:00 am pacific)
6501 Gates Hillman Center
Presented on video at the Univ. of Washington Astro Video Conference Room (B368 PAB)
Jeff Schneider, Carnegie Mellon University
Active Learning for Fitting Astrophysical Simulation Models to Observational Data
SLIDES AVAILABLE - PDF [9.5M]
Researchers in machine learning and operations research have made great
strides in modeling and optimizing manufacturing and other commercial
processes. A more recent trend is to observe that many scientific modeling tasks can be approached with similar techniques. In this talk we consider a specific example of that: using active learning to fit cosmological models based on data from sky surveys. We propose appropriate active learning algorithms, make observations about the implications of using Bayesian vs frequentist methods to test model fits, and address active learning for data/evidence fusion. Results will be presented for the CMBFAST model and WMAP data, and future plans will be described in cosmology and other scientific applications. We will conclude with some observations about active learning in parallel computing architectures.
Speaker Bio:
Dr. Jeff Schneider is an associate research professor in the Carnegie
Mellon University School of Computer Science. He received his PhD in
Computer Science from the University of Rochester in 1995. He has over 15
years experience developing, publishing, and applying machine learning
algorithms in government, science, and industry. He has dozens of
publications and has given numerous invited talks and tutorials on the
subject.
More Information:
Contact
or