James Sharpnack

Postdoctoral Researcher

Mathematics Department at UC San Diego

Ph.D. in Machine Learning and Statistics from Carnegie Mellon University

Email: jsharpna@gmail.com


RESEARCH: The prevalence of massive datasets is transforming the field of statistics as we attempt to simultaneously address the statistical efficiency and computational complexity of statistical methods. The modern explosion of data, viz. internet data, sensor networks, genomic microarrays, and financial data, requires that statistical methodology adapts to the assumptions and structure unique to these problems, while ensuring statistical and computational soundness. My research focuses on the development of algorithms with provably low statistical risk and low computational complexity to new statistical problems with complex structural and modeling assumptions.

My Ph.D. advisors were Aarti Singh and Alessandro Rinaldo, while my postdoc advisor is Ery Arias-Castro

THESIS PROPOSAL: Graph Structured Statistical Inference

DISSERTATION: Graph Structured Normal Means Inference

BIO: I am currently a postdoctoral researcher at UCSD in the Mathematics department. I hold a Ph.D in Machine Learning and Statistics from Carnegie Mellon University. I have a masters in Statistics from CMU and bachelors in Math and Physics from the Ohio State University.


PUBLICATIONS:
Near-optimal Anomaly Detection in Graphs using Lovasz Extended Scan Statistic (arXiv)
J. Sharpnack, A. Krishnamurthy, and A. Singh
Neural Information Processing Systems (NIPS), 2013.
Detecting Activity in Graphs with Spectral Scan Relaxations (arXiv)
J. Sharpnack, A. Rinaldo, and A. Singh
Submitted for publication, 2013.
Near-optimal and Computationally Efficient Detectors for Weak and Sparse Graph-structured Patterns (pdf)
J. Sharpnack and A. Singh
IEEE Global Conference on Signal and Information Processing (GlobalSIP), 2013
A Path Algorithm for Localizing Anomalous Activity in Graphs (pdf)
J. Sharpnack
IEEE Global Conference on Signal and Information Processing (GlobalSIP), 2013
Recovering Graph-Structured Activations using Adaptive Compressive Measurements (arXiv)
A. Krishnamurthy, J. Sharpnack, and A. Singh
Best Student Paper Award
Asilomar Conference on Signals, Systems, and Computers, 2013
Detecting Activations over Graphs using Spanning Tree Wavelet Bases (arXiv)
J. Sharpnack, A. Krishnamurthy, and A. Singh
With Oral Presentation
International Conference on Artificial Intelligence and Statistics (AIStats), 2013
Changepoint Detection over Graphs with the Spectral Scan Statistic (arXiv)
J. Sharpnack, A. Rinaldo, and A. Singh
International Conference on Artificial Intelligence and Statistics (AIStats), 2013
Variance Funtion Estimation in High-Dimensions (pdf)
M. Kolar, and J. Sharpnack
With Oral Presentation
International Conference of Machine Learning, ICML 2012
Sparsistency of the Edge Lasso over Graphs (pdf)
J. Sharpnack, A. Rinaldo, and A. Singh
International Conference on Artificial Intelligence and Statistics (AIStats), 2012.
Identifying Graph-structured Activation Patterns in Networks (pdf)
J. Sharpnack, and A. Singh
With Oral Presentation
Neural Information Processing Systems (NIPS), 2010