Self-checking algorithm interprets gravitational-wave data
Together with an interdisciplinary team from the Max Planck Institute for Intelligent Systems and the Max Planck Institute for Gravitational Physics, including Hector Fellow Bernhard Schölkopf, Young Researcher Maximilian Dax published a paper on Neural Importance Sampling for Rapid and Reliable Gravitational-Wave Inference in the journal Physical Review Letters.
In the publication, he developed an algorithm that immediately checks its own calculations of merging black holes’ properties and corrects its result if necessary – inexpensively and rapidly. The machine learning method provides very accurate information about the observed gravitational waves and will be ready for use when the global network of gravitational-wave detectors starts its next observing run in May.
The developed self-checking deep learning system extracts information from gravitational-wave data very accurately. In the process, the system checks its own predictions about the parameters of merging black holes – a deep neural network with a safety net. A set of 42 detected gravitational waves from merging black holes were successfully analyzed by the algorithm: When cross-checked against computationally expensive standard algorithms, the results were indistinguishable.