Logo
ScienceStack
Table of Contents
Fetching ...
Sign In

A Brief History of Inference in Astronomy

Rafael S. de Souza, Emille E. O. Ishida, Alberto Krone-Martins

Abstract

In this short review, we trace the evolution of inference in astronomy, highlighting key milestones rather than providing an exhaustive survey. We focus on the shift from classical optimization to Bayesian inference, the rise of gradient-based methods fueled by advances in deep learning, and the emergence of adaptive models that shape the very design of scientific datasets. Understanding this shift is essential for appreciating the current landscape of astronomical research and the future it is helping to build.

A Brief History of Inference in Astronomy

Abstract

In this short review, we trace the evolution of inference in astronomy, highlighting key milestones rather than providing an exhaustive survey. We focus on the shift from classical optimization to Bayesian inference, the rise of gradient-based methods fueled by advances in deep learning, and the emergence of adaptive models that shape the very design of scientific datasets. Understanding this shift is essential for appreciating the current landscape of astronomical research and the future it is helping to build.
Paper Structure (7 sections, 25 equations, 6 figures)

This paper contains 7 sections, 25 equations, 6 figures.

Table of Contents

  1. Linear Regression in Astronomy
  2. Measuring the Universe Expansion Rate
  3. Amortized Inference
  4. Active Inference in Cosmology
  5. Remarks

Figures (6)

  • Figure 1: Left: A blurred, Poisson-noised image of the Whirlpool Galaxy M51. Right: Denoised version obtained via the Richardson–Lucy deconvolution. The method enhances the contrast of fine details such as spiral arms. Credit: NASA, ESA/Hubble, and the Hubble Heritage Team (STScI/AURA).
  • Figure 2: Original plot of radial velocity-distance relation for extragalactic nebulae hub29.
  • Figure 3: Hamiltonian Monte-Carlo trajectory.
  • Figure 4: A diagram outlining our amortized inference procedure.
  • Figure 5: Illustration of a normalizing flow transforming a simple base distribution into a complex target distribution through successive invertible mappings.
  • ...and 1 more figures