Emille E. O. Ishida, PhD

I am a Brazilian physicist  working in Astronomy and Cosmology since I can remember.  

My research is focused in type Ia Supernova (photometric classification, characterization and cosmological application), machine learning, Bayesian modeling and the connections between them.

I am also co-founder of the Cosmostatistics Initiative (COIN) and coordinator of their Python-related projects

Main scientific projects

Fink
An adaptable LSST community broker based on machine learning
RESSPECT
REcommendation System for SPECTroscopic follow-up
SNAD
SuperNova Anomaly Detection

MY LATEST RESEARCH

Active Anomaly Detection for time-domain discoveries

Ishida et al., 2019

We present the first application of adaptive machine learning to the identification of anomalies in a data set of non-periodic astronomical light curves. The method follows an active learning strategy where highly informative objects are selected to be labelled. This new information is subsequently used to improve the machine learning model, allowing its accuracy to evolve with the addition of every new classification. For the case of anomaly detection, the algorithm aims to maximize the number of real anomalies presented to the expert by slightly modifying the decision boundary of a traditional isolation forest in each iteration.  As a proof of concept, we apply the Active Anomaly Discovery (AAD) algorithm to light curves from the Open Supernova Catalog and compare its results to those of a static Isolation Forest (IF). For both methods, we visually inspected objects within  2%  highest anomaly scores. We show that AAD was able to identify 80% more true anomalies than IF. This result is the first evidence that AAD algorithms can play a central role in the search for new physics in the era of large scale sky surveys.

Machine Learning methods will play a fundamental role in our ability to optimize the science output from the next generation of large scale surveys. Given the peculiarities of astronomical data, it is crucial that algorithms are adapted to the data situation at hand. In this comment, I review the recent efforts towards the development of automatic systems to identify and classify supernova with the goal of enabling their use as cosmological standard candles.

Laboratoire de Physique de Clermont - LPC

Universite Clermont Auvergne

Clermont-Ferrand, France

  • Facebook Clean Grey
  • Twitter Clean Grey