Due to their powerful information processing capabilities, artificial intelligence (AI) and, more broadly, neural network-based computing are generating increasing interest for their applications in the quantum domain. In two of our recent studies, we demonstrate how quantum optics and these new computational approaches can complement each other.
In the first study, we use artificial intelligence to establish the presence of quantum entanglement in the light generated by our experiment. The preparation of quantum entangled states is a crucial ingredient in quantum technologies. In our experiments, information is encoded in the phase and amplitude of the electric field of a light beam. Unlike qubits, which are often considered for quantum computing, these degrees of freedom are continuous, making the characterization of their entanglement more complex. We designed an approach based on an artificial neural network capable of recognizing whether a quantum state of light is entangled or not, based on the correlations between its continuous degrees of freedom. This pioneering work opens up a new field of application for AI in quantum technology and enriches the available tools for characterizing the quantum resources of light.
In the second study, we implement a machine learning algorithm known as reservoir computing, by encoding information in a network of light pulses. Reservoir computing is an algorithmic approach that, among other things, allows the extrapolation of time series into the future. We designed an experiment implementing such an algorithm, leveraging the multimode structure of laser pulses. We plan to explore the specific quantum resources that could offer an advantage to our optical system compared to classical implementations of the same computation.
These results stem from two successful collaborations between the Multimode Quantum Optics team and Qiongyi He’s group at Peking University in China, and Roberta Zambrini’s group at IFISC (CSIC-UIB) in Spain.
For more information:
- Gao, M. Isoard, F. Sun, C. E. Lopetegui, Y. Xiang, V. Parigi, Q. He, and M. Walschaers, Correlation-pattern-based Continuous-variable Entanglement Detection through Neural Networks, Phys. Rev. Lett. 132, 220202 (2024)
https://journals.aps.org/prl/abstract/10.1103/PhysRevLett.132.220202 - Henaff, M. Ansquer, M. C. Soriano, R. Zambrini, N. Treps, and V. Parigi, Optical phase encoding in a pulsed approach to reservoir computing Opt. Lett. 49, 2097-2100 (2024)
https://opg.optica.org/ol/fulltext.cfm?uri=ol-49-8-2097&id=548914