Overview

Massive non-Euclidean data sets arise ubiquitously in a variety of scientific settings, either naturally in a geometry-free way as relational data amongst entities, or as a result of modern, graph-based, locally linear embeddings designed to achieve dimensionality reduction. In turn, inference for graphs and networks represents a critical growth area for twenty-first century data analysis. Work at SISL touches on random matrix theory, high-dimensional covariance estimation, spectral methods, and statistical inference for network-valued data sets.

Sponsorship

Work sponsored by the Defense Advanced Research Projects Agency (DARPA) and the National Science Foundation (NSF).

Team

PI: Patrick J. Wolfe

Postdoctoral Fellows: Mohamed-Ali Belabbas, David Choi, Patrick O. Perry

Graduate Students: Nicholas Arcolano, Jing Gu, Frank Tompkins, Yuhang Wang

Publications

B. Olding and P. J. Wolfe. "Inference for graphs and networks: Extending classical tools to modern data." Submitted for publication, 2009.

M.-A. Belabbas and P. J. Wolfe. "On landmark selection and sampling in high-dimensional data analysis." Philosophical Transactions of the Royal Society, Series A, vol. 367, pp. 4313-4337, 2009. arXiv:0906.4582

M.-A. Belabbas and P. J. Wolfe. "Spectral methods in machine learning and new strategies for very large data sets," Proceedings of the National Academy of Sciences of the USA, vol. 106, pp. 369-374, 2009.

M.-A. Belabbas and P. J. Wolfe. "On sparse representations of linear operators and the approximation of matrix products," in Proceedings of the 42nd Annual Conference on Information Sciences and Systems, 2008, pp. 258-263.

D. N. Spendley and P. J. Wolfe, “Adaptive beamforming using fast low-rank covariance matrix approximations,” in Proceedings of the IEEE Radar Conference (RadarCon), 2008, in press.

M.-A. Belabbas and P. J. Wolfe, “Fast low-rank approximation for covariance matrices,” in Proceedings of the 2nd IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP), 2008, pp. 293–296.

F. Tompkins and P. J. Wolfe, “Bayesian filtering on the Stiefel manifold,” in Proceedings of the 2nd IEEE International Workshop on Computational Advances in Multi-Sensor Adaptive Processing (CAMSAP), 2008, pp. 261–264. Special Session on Monte Carlo Methods for Multisensor Processing.

R. K.-X. Jin and D. C. Parkes and P. J. Wolfe, “Analysis of Bidding Networks in eBay: Aggregate Preference Identification through Community Detection,” in Plan, Activity, and Intent Recognition (PAIR): Papers from the 2007 AAAI Workshop, C. Geib and D. Pynadath, Eds., pp. 66–73. AAAI Press, Menlo Park, 2007, Technical Report WS-07-09.

J. Gu and P. J. Wolfe, “Robust adaptive beamforming using variable loading,” in Proceedings of the 4th IEEE Workshop on Sensor Array and Multichannel Signal Processing (SAM), 2006, pp. 1–5.

P. Parker, P. J. Wolfe, and V. Tarokh, “A signal processing application of randomized low-rank approximations,” in Proceedings of the 13th IEEE Workshop on Statistical Signal Processing (SSP), 2005, pp. 345–350.