Machine learning ("artificial intellegence") improves experiment

21 June 2023

Seismic disturbances are a problem in many physical experiments, including gravitational wave observatories. We were able to build an artificial neural network that has learnt from past disturbances and is better able to recognise similar acute disturbances.

Nicolas Heimann, Jan Petermann, Daniel Hartwig and Ludwig Mathey, Predicting the motion of a high-Q pendulum subject to seismic perturbations using machine learning , Applied Physics Letters 122 (June 2023)

Latest articles

25.08.2025|IQP Schnabel

Generation of Schrödinger cat states

By subtracting a photon from squeezed vacuum states, it is possible to generate (microscopic) Schrödinger cat states

In our publication, Optica Quantum, Vol. 3, No 4/25 August 2025, we achieved for the first time (with a certain probability) the modification of the corresponding measurement data to...

05.08.2025|IQP Schnabel

Generation and explanation of the Einstein-Podolsky-Rosen entanglement

Is it possible to predict measurement results without being limited by the Heisenberg uncertainty relation? Yes, but only if you use “EPR” entanglement. There are already various publications on this topic from the working group, e.g., [Zander et al. (2021)].

With his new work Roman Schnabel...