Shoubaneh HEMMATI
(Caltech/IPAC)
We introduce a new technique for measuring physical properties of galaxies (integrated and resolved) using neural networks machine learning Self Organizing Maps (SOMs) to replace typical SED-fitting techniques. We demonstrate how SOMs can be used to visualize and optimize libraries of stellar evolution synthesis models. We will present comparison of our SOM measured properties, such as stellar masses and star formation rates of COSMOS galaxies to the previous published measurements. In addition, we show how SOMs are able to provide a much quicker measurement of the galaxy properties, which is essential for processing the massive datasets expected from next generation galaxy surveys, such as WFIRST.