Friday, April 22, 2016

Announcement: The Science Circle - “Measuring the Universe With Machine Learning Methods”

By Giuseppe Longo – Giulon Ragu

10 AM SL Time
Modern digital sky surveys produce terabytes of high quality data which can be used to achieve a panchromatic view of the Universe of unprecedented depth and complexity.

Hundreds of millions of galaxies are observed and the parameters describing their morphological and spectral properties are stored in large databases which are often accessible to the astronomical community at large.

One of the key issues to pursue the scientific goals of these surveys is to estimate with accuracy the distance (or the redshift) of these galaxies: a fact which cannot be done using traditional spectroscopic techniques but requires advanced methods capable to estimate the redshifts using photometric data alone.

A difficult task which can be achieved only exploiting the possibilities offered by the modern computer infrastructures and methods derived from the machine learning methodology.

The talk will outline the main problem of evaluating galaxy distances and will focus on the most modern methods used to derive the so called “photometric redshifts”.

The Science Circle (139/206/95)

No comments:

Post a Comment