Machine Learning for Quantum Optics

This theoretical beam is created by deep learning to efficiently perform denoising of optical data. (Source: Skoltech / NPG)

As machine learning continues to surpass human performance in a growing number of tasks, scientists at Skoltech have applied deep learning to reconstruct quantum properties of optical systems. Through a colla­boration between the quantum optics research labora­tories at Moscow State University, led by Sergey Kulik, and members of Skoltech’s Deep Quantum Laboratory of CPQM, led by Jacob Biamonte, the scientists have success­fully applied machine learning to the state recon­struction problem.

Their findings are the first to show that machine learning can reconstruct quantum states from experimental data in the presence of noise and detector errors. Skoltech PhD student Adriano Macarone Palmieri described the findings as  “a new open door towards deeper insights.” Federico Bianchi, a machine learning expert, describes the findings as “a sound example of data driven disco­very which combines machine learning and quantum physics.” While Federico didn’t have experience with quantum mechanics prior to joining this study, he viewed the problem in terms of infor­mation and helped create a novel model of the system based on deep feed forward neural networks.

Both Adriano and Federico worked tirelessly and in close colla­boration with many members of Deep Quantum Laboratory, including Dmitry Yudin who describes the findings as an important first step towards the practical use of neural network architecture in a lab for improving quantum tomo­graphy with available quantum setups of noisy experi­mental data. Such quantum information processing is used ubiquitously in paradigmatic quantum devices for quantum computation and optimi­zation. In the forthcoming future, the researchers plan to address further challenges of upscaling quantum information devices, and expect this work to be founda­tional in their further research.

These results wouldn’t have been possible without the experimental research of Egor Kovlakov, supported by Stanislav Straupe and Sergei Kuliik, from MSU. Over the last several years, they have applied a wide range of techniques to the state recon­struction problem. To their surprise, deep learning out­performed these state-of-the-art methods in a real experiment. The MSU team generated data with an experi­mental platform based on spatial states of photons to prepare and measure high-dimensional quantum states. Experimental errors in state preparation and measurements inevitably plague the results and the situation becomes worse with increasing dimen­sionality.

At the same time, extending the dimen­sionality of accessible quantum states is extremely important for quantum communi­cation protocols and, especially, quantum computing. This is where machine learning techniques come in useful. The Skoltech team imple­mented a deep neural network implemented to analyze the noisy experimental data and effi­ciently learn to perform denoising, signi­ficantly improving the quality of quantum state recon­struction.

Skoltech’s Deep Quantum Labora­tory team believes that machine learning techniques will play an essential role in the future development of quantum tech­nologies. As the available quantum devices become more and more complex, it gets harder and harder to control all the parameters at the desired level of precision. This came out as a very natural field of appli­cation for deep learning and machine learning techniques in general. (Source: Skoltech)

Reference: A. M. Palmieri et al.: Experimental neural network enhanced quantum tomography, npj Quant. Infor. 6, 20 (2020); DOI: 10.1038/s41534-020-0248-6

Link: Skolkovo Institute of Science and Technology, Moscow, Russia

Speak Your Mind

*