What is it about?
Our study delves into the fascinating world of gravitational lensing, a phenomenon where massive celestial objects warp the fabric of space and time. Gravitational lensing provides valuable insights into fundamental physics and lets us observe otherwise hidden astronomical entities. In recent years, we have witnessed the power of machine learning in deciphering gravitational lensing effects. However, existing methods have limitations in analysing complex datasets that combine images and time-series data. This paper introduces DeepGraviLens, an innovative artificial intelligence approach that can accurately classify spatio-temporal data associated with non-lensed and lensed systems.
Featured Image
Photo by Conner Baker on Unsplash
Why is it important?
We created DeepGraviLens, a new artificial intelligence approach that outperforms other methods and can enhance the efficiency of studying lensed objects in upcoming astrophysical surveys, utilising massive amounts of data. Our approach has shown good results both on simulated data and in real data.
Read the Original
This page is a summary of: DeepGraviLens: a multi-modal architecture for classifying gravitational lensing data, Neural Computing and Applications, June 2023, Springer Science + Business Media,
DOI: 10.1007/s00521-023-08766-9.
You can read the full text:
Contributors
The following have contributed to this page
