Ecological processes such as bird migration are complex, difficult听to observe, and occur at the scale of continents, making it impossible for听humans to grasp their听broad-scale patterns directly. However, novel data听sources鈥攕uch as large sensor networks and millions of bird observations听reported by human 鈥渃itizen听scientists"鈥攁re providing new opportunities to听understand ecological phenomena at very large scales. The ability to fit听models, test hypotheses, make predictions,听and reason about human impacts on听biological processes at this scale promise to revolutionize ecological science听and environmental policy.
听
In this talk, I will present novel algorithmic approaches to听overcome challenges throughout the 鈥減ipeline鈥 from low-level data听interpretation to model fitting to听high-level decision-making in large-scale听ecological science, including: (1) biological interpretation of NEXRAD weather听radar, (2) probabilistic modeling of bird听migration using citizen science data听and (3) optimizing land purchases to support the recovery of endangered听species. I will highlight contributions from this work听that extend well beyond听ecology, including a very general optimization framework for maximizing the听spread of a cascading process in a network, and a formalism听called听Collective Graphical Models听for听efficiently reasoning about probabilistic models of large populations of听individuals when only aggregate data is available.
Daniel Sheldon is a postdoctoral fellow in the School of EECS at 精东影视 State University, where he holds an NSF fellowship in Bioinformatics. His primary research interests are machine learning and probabilistic modeling applied to large-scale problems in ecology and computational sustainability. Other research interests include web search and reputation systems, optimization, statistics, and network modeling. He completed his Ph.D. in computer science at Cornell University in 2009. Prior to that, he received an A.B. in mathematics from Dartmouth College in 1999, and worked at Akamai Technologies and then DataPower Technology between 1999 and 2004.