top of page
White Structure

Galactic Triumphs: A Deep Learning Approach for Decoding Temperature in Galaxy clusters

Writer's picture: Asif Iqbal AhangarAsif Iqbal Ahangar








Fig. 1: Comparison between learning in the standard Auto-Encoder and IAE models. Note that after training, the encoder is not needed anymore and the decoder (IAE model) acts as a temperature model which is the function of the barycentric weights. See text for more details.





Fig. 2: The left panel shows the reconstructed 3D temperature profile obtained by directly fitting the 3D temperature input profiles to the IAE model and the right panel shows the reconstructed 3D temperature profiles by fitting 2D temperature profiles to the IAE models. Note that over most of the radial range, the difference between the true and recovered 3D temperature from the IAE model is less than 10% over most of the radial range.

Comments


  • download
  • Facebook Clean Grey
bottom of page