SEMINÁRIO DO DEPARTAMENTO DE ASTRONOMIA
Unsupervised Methods in Morphological Classification Applied to Galaxy Evolution
by Prof. Reinaldo Ramos de Carvalho (Universidade Cidade de São Paulo)
Abstract:
Galaxy morphology offers significant insights into the evolutionary pathways and underlying physics of galaxies. As astronomical data grow with surveys such as CEERS/JWST, there is a need for tools to classify and analyse the vast numbers of galaxies that will be observed. In this presentation I will introduce a novel classification technique blending unsupervised clustering based on morphological metrics with the scalability of supervised Convolutional Neural Networks. We present a comparative analysis between the well-known CAS (Concentration, Asymmetry, and Smoothness) metrics and our newly proposed EGG (Entropy, Gini, and Gradient Pattern Analysis). Our choice of the EGG system stems from its separation-oriented metrics, maximizing morphological class contrast. Applying our methodology to the Sloan Digital Sky Survey data, we obtain ∼95% of Overall Accuracy of purely unsupervised classification and when we replicate T-Type and visually classified galaxy catalogues with accuracy of ∼88% and ∼89%, respectively. Furthermore, the application to Hubble Space Telescope data heralds the potential for unsupervised exploration of a higher redshift range. A notable achievement is our ∼95 per cent accuracy in unsupervised classification, a result that rivals when juxtaposed with Traditional Machine Learning and closely trails when compared to Deep Learning benchmarks. Some preliminary results using CANDELS data will be presented and discussed.
Short-Bio:
Possui graduação em Astronomia pela Universidade Federal do Rio de Janeiro (1978), mestrado em Astronomia pelo Observatório Nacional (1983) e doutorado em Astronomia pelo Observatório Nacional (1988). Fez pós-doutorado em Caltech em 1988/1989 trabalhando com o Dr. George Djorgovski. Entre 1995 e 1997 foi Senior Research Fellow em Caltech, participando ativamente do projeto Digitized Palomar Observatory Sky Survey. Atualmente é professor da Pós-Graduação da Universidade Cidade de São Paulo Atua principalmente na área de Cosmologia Observacional com ênfase no estudo das propriedades globais de Aglomerados de Galáxias e nos aspectos ligados à evolução de galáxias. Membro do International Advisory Board do International Institute of Physics. Editor Associado da Frontiers Space Science - Astronomy. É pesquisador com bolsa de produtividade do CNPq nível Senior. Foi Presidente da Sociedade Astronômica Brasileira entre 2017 e 2020.
Google Meet: https://meet.google.com/pcw-gmem-jyi
Link da transmissão: https://www.youtube.com/c/AstronomiaIAGUSP/live