01.10.2023

                 This year’s LISA Astrophysics Working Group Meeting (https://sites.google.com/unimib.it/lisaastrowgmeeting/home?authuser=0 ) took place at the University of Milano-Bicocca in Italy, between the 13th and 15th of September. For three whole days, a lot of LISA science was discussed, as well as possible synergies with other Gravitational Wave Physics related topics. During this event, one of our members gave a presentation on The Impact of Gravitational Waves on Multi-Messenger Analysis of Observed Electromagnetic Sources, whose abstract you can read in the following lines:

               “Astrophysical events, such as mergers of black holes and neutron stars or strong emitting sources such as pulsars or supernovae, can be better understood when analyzed from multiple perspectives. With the current and upcoming gravitational wave experiments, we expect to add gravitational wave (GW) signals as important messengers for analyzing these kinds of events.  In this work, we present our studies on the influence of GW source parameters upon the shape of the waveform and on how we can use these new detected and characterized messengers to improve our knowledge of sources already observed in electromagnetic and with astrophysical particles. We also show the results of our multi-messenger analysis on different astrophysical sources such as quasars and X-shaped radio galaxies.”

      Between 17th and 22nd of September, the IEEE Quantum Week  was held in Bellevue, Washington Hyatt Regency Bellevue on Seattle’s Eastside (https://qce.quantum.ieee.org/2023/).  Our Quantum wizard, also known as Maria, held a presentation entitled “Classification of Gravitational Waves Using Neural Networks on Quantum Computers” within Quantum AI Workshop. She discussed the feasibility of using quantum neural networks for classifying gravitational waveforms, using both simulators and quantum computers. The analysis was quite interdisciplinary in its nature, combining knowledge involving astrophysics, quantum information as well as quantum and classical machine learning. On one hand, we showed that the quantum classifiers and hybrid classical-quantum neural networks give maximally accurate results when tested on a simple dataset and ran on a simulator. On the other hand, we showed that quantum neural networks (ran on a simulator) can distinguish with high accuracy between noisy gravitational waves and noise, for LISA space mission specific simulated data. Moreover, we showed that adding a quantum layer to poorly performing classical neural network can highly improve its accuracy. When running any of our quantum algorithms on a real quantum computer, error minimizing algorithms need to be implemented in order to obtain a satisfying accuracy.