Knowledge of the redshift – or distance – of an astronomical object is vital for many scientific endeavours, from massively large-scale projects like testing Cosmological theories and measuring the Cosmic Star Formation History, to the small, individual projects investigating the properties of single objects, where the observed brightness of the object is often highly dependent on its distance. Radio galaxies are of particular interest as they are typically observed at higher than normal redshift, giving observers a view into the early Universe. However, one of the reasons they are visible at higher redshift – they typically have an active supermassive black hole at the centre of the galaxy – is also the reason a lot of traditional methods for estimating redshift fail. The emission from a black hole is difficult to separate from the emission from the stars forming inside the galaxy hosting that black hole, for example. This work compares simple, traditional machine learning algorithms like the k-Nearest Neighbours (kNN), with much more complex algorithms based around Neural Networks and Gaussian Processes. We find that as long as the available model parameters are optimised, the simple kNN algorithm performs best for our use case, although it does have its limitations. Finally, we are able to estimate the redshift of 100,000 radio sources in the Evolutionary Map of the Universe Pilot Survey, many of which have never been seen before at radio wavelengths.
Hello world!
Welcome to WordPress. This is your first post. Edit or delete it, then start writing!
optimize2023
April 13, 2023
