Buscador de actividades

Por fecha

Búsqueda por rango de fechas

Por unidad

Por espacio

Conferencias y seminarios

Seminario DAS: "Machine-aided anomaly detection in large data sets"


Jueves 06 de Junio de 2019


12:15 hrs.


Sala de Conferencias Federico Ristenpart (Camino El Observatorio 1515, Departamento de Astronomía, Edificio Central, piso 3, Las Condes, Santiago, Chile)

Agregar a calendario

Descargar archivo (ical)
Copiar Enlace
Machine-aided anomaly detection in large data sets

Machine-aided anomaly detection in large data sets

Enlaces relacionados
Departamento de Astronomía

Speaker: Aleksandra Solarz
National Centre for Nuclear Research, Warsaw, Poland

Title: Machine-aided anomaly detection in large data sets

Abstract: Wide-angle photometric surveys of previously uncharted sky areas or wavelength regimes will always bring in unexpected sources whose existence and properties cannot be easily predicted from earlier observations. Such objects can be efficiently sought for with novelty detection algorithms. I will present an application of such a method, called one-class support vector machines (OCSVM), to search for anomalous patterns among sources pre-selected from the mid-infrared AllWISE catalogue covering the whole sky. OCSVM successfully finds artifacts, such as objects with spurious photometry due to blending, but most importantly also real sources of genuine astrophysical interest. Among the latter, OCSVM has identified a sample of heavily reddened AGN/quasar candidates distributed uniformly over the sky and in a large part absent from other WISE based AGN catalogues. It also allows to find a specific group of sources of mixed types, mostly stars and compact galaxies. By combining the semi-supervised OCSVM algorithm with standard classification methods it will be possible to improve the latter by accounting for sources which are not present in the training sample but are otherwise well-represented in the target set.

Rene A. Mendez
Seminar Coordinator
DAS/UChile - rmendez@uchile.cl

Departamento de Astronomía, Cerro Calán
Alejandro Leal Obreque - +562 29771154 -

Departamento de Astronomía