The Large Hadron Collider pioneered quantum machine learning

The Large Hadron Collider (LHC) was relaunched in the spring of 2022 after three years of technical

maintenance and modernization.Immediately after the launch, scientists announced an energy record achieved at the LHC. To take full advantage of the massive increase in data flow, the researchers are using quantum machine learning for jet analysis for the first time.

Quantum machine learning methods are alreadywere used in particle physics to solve problems of classifying events and reconstructing particle tracks, but the team was the first to use them to identify the charge of a hadron jet. To do this, scientists have developed a variational quantum classifier based on two different quantum schemes.

Physicists used a quantum simulator tocompare the effectiveness of the new method and currently used deep neural networks. It turned out that the quantum circuit is still slightly inferior in performance, but the difference is not great.

The performance of various algorithms independing on the transverse momentum of the jet. DNN - traditional deep learning, Angle Emb. and Amplitude Amb. — quantum circuits. Image: Alessio Gianelle et al., Journal of High Energy Physics

At the same time, a new method using quantumnetworks achieve optimal performance with fewer events. This will help reduce the use of resources to process the huge data streams received at the LHC. At the same time, when using a large number of functions, deep machine learning still outperforms quantum algorithms. Scientists believe this will change as more efficient quantum hardware becomes available.

The researchers also found that quantumalgorithms allow studying correlations between functions. This is necessary to extract information about the correlations of the jet components. And, therefore, quantum analysis will improve the identification of the flavor of the hadron jet.

Using quantum machine learning so faris in its infancy, the authors say. As physicists gain experience with quantum computing, radical improvements in hardware and computing technology should be expected.

Read more:

Soon a solar storm will hit the Earth: the material flies at a speed of 800 km / s

Scientists filmed a strange creature with tentacles, which they mistook for a flower

Russia leaves the ISS: what will happen now and why the maintenance of the station is under threat