Home    Data    Software    Courses

Diane Oyen

 

Research Scientist

Los Alamos National Lab


Research Interests

I develop machine learning methods that help scientists to understand complex data.

Education

PhD, Computer Science, University of New Mexico

    Dissertation: Interactive Exploration of Multitask Dependency Networks

    Advisor, Terran Lane

BS, Electrical and Computer Engineering, Carnegie Mellon University

Select Publications

D. Oyen, B. Anderson, and C. Anderson-Cook. Bayesian networks with prior knowledge for malware phylogenetics. In The AAAI-16 Workshop on Artificial Intelligence and Cybersecurity, 2016.

  1. J.K. Johnson, D. Oyen, M. Chertkov, and P. Netrapalli. Learning planar Ising models. Journal of Machine Learning Research, 2016.

D. Oyen, R. B. Porter, and K. Sentz. Interactive comparative analysis for multi-modal data exploitation and fusion. In IEEE Applied Imagery Pattern Recognition (AIPR-15), 2015.

D. Oyen, N. Lanza, and R. B. Porter. Discovering compositional trends in Mars rock targets from ChemCam spectroscopy and remote imaging. In IEEE Applied Imagery Pattern Recognition (AIPR-15), 2015.

R. B. Porter, D. Oyen, and B. G. Zimmer. Learning watershed cuts energy functions. International Symposium on Mathematical Morphology, (2015).

D. Oyen and N. Lanza. Discovering chemical structure in ChemCam targets using Gaussian graphical models: Compositional trends with depth. Lunar and Planetary Institute Science Conference (LPSC), 2015.

J. K. Johnson, D. Oyen, M. Chertkov, and P. Netrapalli. Learning planar Ising models. Technical Report LA-UR-15-20740, arXiv, 2015.

D. Oyen and T. Lane. Transfer learning for Bayesian discovery of multiple Bayesian networks. Knowledge and Information Systems, pp 1–28, 2014.

D. Oyen and T. Lane, Interactive exploration of comparative dependency network learning. KDD Workshop on Interactive Data Exploration and Analytics (IDEA), 2014.

D. Oyen and T. Lane, Bayesian discovery of multiple Bayesian networks via transfer learning. IEEE International Conference on Data Mining (ICDM), 2013.

D. Oyen, A. Niculescu-Mizil, R. Ostroff, A. Stewart, and V. P. Clark. Controlling the precision-recall tradeoff in differential dependency network analysis. ArXiv, 2013.

D. Oyen, A.Niculescu-Mizil, R. Ostroff, and A. Stewart. Controlling the precision-recall tradeoff in differential dependency network analysis. Workshop on Machine Learning in Systems Biology (MLSB), 2013.

D. Oyen and T. Lane. Leveraging domain knowledge in multitask Bayesian network structure learning. AAAI Conference on Artificial Intelligence, 2012.

D. Oyen. Active learning of transfer relationships for multiple related Bayesian network structures. IEEE ICDM Workshops, 2011.

E. Besada-Portas and D. Oyen. Redes Bayesianas. In Aprendizaje Automatico: Un Enfoque Practico, Ra-Ma, 2010.  

Data

Multivariate time series data compiled from the USGS Streamflow database.

http://www.cs.unm.edu/~doyen/data/streamflow/

Software

Open-source code available here.

Courses Taught at UNM

Honors 302-005: How to Lie with Statistics: Uses and Misuses of Numbers in Argument

    Course website

Professional Activities

Board Member & Annual Meeting Chair, New Mexico Network for Women in Science and Engineering
Senior Program Committee Member, AAAI-13
Scholar, National Security Studies Program (NSSP)
past Organizer, Women in Machine Learning Workshop
past Vice President, Computer Science Graduate Student Association
past Program Chair, Computer Science @ UNM Student Conference