Caren Marzban

Caren Marzban, Ph.D.
Principal Physicist, Applied Physics Laboratory,
Lecturer, Department of Statistics, University of Washington

My vitae (out of date).
My courses: AI short course , Stat220, Stat311, Stat427, Stat/Math394, Stat/Math390 Stat421, Stat509.

My son, Ethan.
My wife's work.
My other works: Forged By Evolution.
My patent.

Journal Articles:

Technical Reports and Unpublished/In-Progress Work:

Selected Conference Papers:

  • Marzban, C. 2017: A few meteorological applications of Sparse PCA. Paper presented at the 16th Conference on Artificial Intelligence, Baltimore, MD, 28-29 July.

  • Marzban, C., R. Tardif, S, Sandgathe, C. Jones, X. Du, N. Li, N. Hryniw, N. C. Lederer, J. D. Doyle, Y. Jin, 2017: The effect of model parameters on the spatial structure of forecast fields. Paper presented at the 24th Conference on Probability and Statistics, Baltimore, MD , 28-29 July.

  • Marzban, C., and R. Viswanathan, 2017: On the Complexity of Neural-Network-Learned Functions. Paper presented at the 15th Conference on Artificial Intelligence, at the 97th American Meteorological Society Annual Meeting, Seattle, Jan. 22-26.

  • Marzban, C., and U. Yurtsever, 2017: On the Shape of Data. Paper presented at the 15th Conference on Artificial Intelligence, at the 97th American Meteorological Society Annual Meeting, Seattle, Jan. 22-26.

  • Marzban, C., and U. Yurtsever 2011: Baby Morse theory in data analysis. Paper at the workshop on Knowledge Discovery, Modeling and Simulation (KDMS), held in conjunction with the 17th ACM SIGKDD Conference on Knowledge Discovery and Data Mining, San Diego, CA., August 21-24.

  • Marzban, C. 2011: Sensitivity Analysis in linear and nonlinear models: A review. Paper presented at the 9th Conference on Artificial Intelligence, at the 91st American Meteorological Society Annual Meeting, Seattle, Jan. 23-27.

  • Marzban, C. 2008: Quantile Regression. Invited paper presented at the joint session between AI and Prob & Stats Conference. 88th American Meteorological Society Annual Meeting, New Orleans, Jan. 20-24.

  • Marzban, C., S. Sandgathe, and H. Lyons 2007: Assessment of an automatic, object-oriented approach to the verification of spatial fields . Paper presented at 7th Euopean Meteorological Society Annual MeetingEl Escorial, Spain, October.

  • Marzban, C. 2004: Probabilistic Forecasts in Meteorology. Talk presented at a Neural Information Processing Systems, 2004, workshop on Calibration and Probabilistic Prediction in Supervised Learning. Whistler, Canada.

  • Marzban, C. 1998: Bayesian inference in neural networks. 78th meeting of the American Meteorological Society, Probability and Statistics Session, Phoenix Arizona, January.

  • Marzban, C., G. J. Stumpf, 1996: A Neural Network for Tornado and/or Severe Weather Prediction Based on Doppler Radar-derived Attributes. 10th Annual Mid-American Symposium on Emerging Computer Technologies, University of Oklahoma, October 28-29. (Top-paper Award.)

  • Marzban, C., R. Viswanathan, 1993: Stochastic Neural Networks and the Weighted Hebb Rule. Proceedings of the IJCNN conference, Nagoya, Japan.


Artificial Intelligence Methods in the Environmental Sciences, 2008; Springer-Verlag. Co-editor and contributor to 2 chapters.

How to Contact me:

Dept. of Statistics
University of Washington
Box 354322
Seattle, WA 98195-4322
Tel: 206.543.7237
Fax: 206.685.7419
marzban at

Applied Physics Laboratory
University of Washington
Box 355640
Seattle, WA 98105-6698
Tel: 206.221.4361
Fax: 206.543.1301
marzban at