Gravitational wave
Ripples in spacetime produced by accelerating massive objects — detected by laser interferometers and increasingly analyzed with AI to extract source parameters from noisy signals.
Gravitational waves are distortions in spacetime caused by the acceleration of massive objects, predicted by Einstein’s general theory of relativity and first directly detected in 2015 by LIGO (merging black holes, ~1.3 billion light-years away). Their detection marked the beginning of gravitational wave astronomy as an observational field.
The signals are extraordinarily faint — the spacetime distortion at Earth is smaller than 1/1000th the diameter of a proton. Detecting them requires kilometer-scale laser interferometers (LIGO, Virgo, KAGRA) and sophisticated signal processing to distinguish the wave from instrumental and environmental noise.
Where AI enters:
Gravitational wave science involves two tasks well-suited to machine learning:
-
Detection and classification: identifying gravitational wave signals buried in noisy interferometer data. Neural networks can scan data streams faster than matched-filter searches and can flag signal candidates for human review.
-
Parameter estimation: determining the source properties — masses, spins, distance, sky location — from the detected signal. Classical Bayesian inference using nested sampling (the standard approach) can take hours to days per event. Simulation-based inference methods trained on synthetic signals can compress this to seconds.
Significance for AI-in-science research:
Gravitational wave analysis is one of the clearest examples of AI accelerating a workflow in physics without replacing the underlying science: physicists still define what parameters to infer; AI speeds up the computation that was previously a bottleneck. It’s a useful case study in how ML fits into a physically grounded workflow.
Related terms: Simulation-Based Inference
Related guide: Physics & Astronomy