A multivariate approach to decode star formation histories from galaxy spectra.

One of the key aspects of extragalactic astrophysics involves the determination of the past star formation history of galaxies. This
allows us to backtrack the formation process, to understand how galaxies form in relation to intrinsic properties -- such as mass or
size -- or to the environment where they live. This PhD project focuses on improving over traditional techniques aimed at extracting
absorption line indicators from galaxy spectra, which are dependent on the age and metallicity distribution of the underlying, unresolved
stellar populations. The methodology involves new multi-variate techniques that optimally extract information from a large volume of
data (typically from thousands of galaxy spectra). This timely extension beyond standard measurements of stellar populations will
emphasize on the NUV spectral range, which -- in combination with optical information -- will enable us to disentangle the contributions
from different stages of the star formation history. The NUV is especially important as very distant, high redshift galaxies, can only
be observed in detail in their spectral NUV rest frame.