CFP: http://www.nsf.gov/funding/pgm_summ.jsp?pims_id=504687
Date: January 23, 2012
Growing out of scientific computation and the explosion in production of digital and observational data, Computational and Data-Enabled Science and Engineering (CDS&E, http://www.nsf.gov/mps/cds-e/) is clearly emerging as a distinct intellectual and technological discipline lying at the interface of mathematics, statistics, computational science, core sciences and engineering disciplines. CDS&E, broadly interpreted, now affects virtually every area of science and technology, revolutionizing the way science and engineering are done.
The Division of Mathematical Sciences and the Office of Cyberinfrastructure of the National Science Foundation recognize the importance of research in CDS&E and envision that the mathematical and statistical research communities will play a leading role in the future development of this emerging science. In partnership with the Office of Cyberinfrastructure, the CDS&E program in DMS supports fundamental research at the core of this emerging discipline. It supports broadly innovative, ambitious and transformative research that will lead to significant advancement in CDS&E. The emphasis will be on mathematical, statistical, computational, and algorithmic developments, as well as their applications in advancing modern cyberinfrastructure and scientific discovery. Multidisciplinary collaboration and the training of the next generation data and computational scientists firmly grounded and trained in mathematics and statistics will be strongly encouraged. The research topics supported by CDS&E -MSS will be rooted in mathematics and statistics and will address computational and big data challenges and promote directly discoveries and innovations at the frontiers of science and engineering. The overall impact in the mathematical and statistical sciences of the proposed work will be a review criterion.
Examples in which mathematical and statistical research enables advances in CDS&E include, but are not limited to:
• Sophisticated computational/statistical modeling for simulation, prediction, and assessment in large scale and data intensive scientific problems that incorporate high performance and/or distributed computing that includes addressing challenges of scalability and heterogeneous architectures
• State-of-the-art tools and theory in statistical inference, statistical learning and data mining for knowledge discovery from massive, complex, and dynamic data sets; or novel usage of knowledge in science to understand effective ways to exploit massive and quickly growing data
• General theory and algorithms for advancing large-scale modeling for complex problems such as those with strong heterogeneities and anisotropies, multi physics coupling, multiscale behavior, stochastic forcing, uncertain parameters or dynamic data, and the subtle impact on a calculation of long-time integration
• Sophisticated computational methods for the elucidation of topological theory, revealing and examining structures in algebraic and arithmetic geometry and number theory, and design of cryptographic security and cybersecured systems
• Innovative methodologies and theory for large scale data acquisition through optimal designs, complex computer experiments, and compressed sampling.
• Study of mathematical, statistical and stochastic properties of complex networks arising from computational science, all other core sciences, and engineering disciplines that are supported by NSF
• Computational differential geometry for graphics and visualization, signal processing, analysis and compressed sensing.
• Advances in discretization methods and solvers, optimization, validation and uncertainty quantification, and automated and reproducible science through rigorous problem specification and code generation
