Model Architect & Software Developer
I build models to solve problems
My work is taking on difficult questions and developing computer simulations and models to answer them. A researcher, I build all sorts of models: simple and large, stochastic and deterministic, conceptual and predictive. Each system and need is different.
I develop custom algorithms, software and interfaces
Off-the-shelf software is great and I use it whenever I can. However, some challenges are just too complex for existing solutions. My work encompasses the entire life-cycle of model construction: from developing special methods to make the best use of the data, to programming the algorithms and model, to developing visualizations and designing accessible interfaces.
Accurately Measuring Model Prediction Error
When assessing the quality of a model, being able to accurately measure its prediction error is of key importance. Often, however, techniques of measuring error are used that give grossly misleading results. This can lead to the phenomenon of over-fitting where a model may fit the training data very well, but will do a poor job of predicting results for new data not used in model training. Here is an overview of methods to accurately measure model prediction error. Read →
Understanding the Bias-Variance Tradeoff
When we discuss prediction models, prediction errors can be decomposed into two main subcomponents we care about: error due to "bias" and error due to "variance". There is a tradeoff between a model's ability to minimize bias and variance. Understanding these two types of error can help us diagnose model results and avoid the mistake of over- or under-fitting. Read →
Publications and Talks
- Fortmann-Roe S (2013). A New Method for the more Accurate Measurement, Communication and Comparison of Model Errors. Transatlantic Conference on Policy Modeling. Washington DC.
- Marston D, Mesick C, Hubbard A, Stanton D, Fortmann-Roe S, Tsao S & Heyne T (2012). Delta Flow Factors Influencing Stray Rate of Escaping Adult San Joaquin River Fall-Run Chinook Salmon (Oncorhynchus tshawytscha). San Francisco Estuary and Watershed Science. 10(4).
- Fortmann-Roe S, Starfield R, & Getz WM. (2012). Contingent Kernel Density Estimation. PLoS ONE, 7(2).
- Nathan R, Spiegel O, Fortmann-Roe S, Harel R, Wikelski M, & Getz WM (2012). Using tri-axial acceleration data to identify behavioral modes of free-ranging animals: general concepts and tools illustrated for griffon vultures. Journal of Experimental Biology, 215(6).
- Fortmann-Roe S, Nathan R, Spiegel O, Harel R and Getz WM (2011). Machine Learning Prediction of Vulture Behavior Using Accelerometer Data. Conference of the European Society for Mathematical and Theoretical Biology. Poland.
- Fortmann-Roe S. (2011). Effects of hue, saturation, and brightness on color preference in social networks: Gender-based color preference on the social networking site Twitter. Color Research and Application.
- Fortmann-Roe S, Iwanejko R, & Wójcik W. (2009) A proposal for a dynamic risk assessment method. Information Systems Architecture and Technology: IT Technologies in Knowledge Oriented Management Process. Ed. Zofia Wilimowska, et al. Biblioteka Informatyki Szkół Wyższych; Wroclaw, pp 17-26. Poland.
- Fortmann-Roe S & Wójcik W. (2009) Symulacja oraz zarządzanie: aplikacje do optymalnego poziomu dezynfekcji Cryptosporidium. Aktualne zagadnienia w uzdatnianiu i dystrybucji wody 1, 331-340. Poland.
- Getz WM, Fortmann-Roe S, Cross P, Lyons A, Ryan S, & Wilmers, C. (2007). LoCoH: nonparameteric kernel methods for constructing home ranges and utilization distributions. PLoS ONE, 2(2), 207.
|National Science Foundation Graduate Research Fellowship |
2012 - 2015
|Berkeley Fellowship |
2010 - 2012
|National Defense Science and Engineering Graduate Fellowship |
2010 (Declined due to overlap)
|Best Paper ISAT Conference |
|Fulbright Fellowship |
2008 - 2009
Featured Modeling Software
Featured Modeling Software
Here are two.