Galaxy Classification in the Perseus Cluster With a Convolutional Neural Network

Type: Poster
SessionPosters (Wednesday & Thursday)
Author: Jason Pruitt

Abstract: We apply the deep learning architecture ResNet-50 trained on Sloan Digital Sky Survey image thumbnails labeled from a combination of bright (r < 19.4) tagged Perseus Cluster data and existing spectroscopic data (0.01 < z < 0.033) to classify galaxies by cluster membership. On an independent test set we find that our method achieves promising results on relevant classification metrics.

Career Stage: 
Grad Student

User login

CAPTCHA
This question is for testing whether or not you are a human visitor and to prevent automated spam submissions.
10 + 1 =
Solve this simple math problem and enter the result. E.g. for 1+3, enter 4.