DeepQuark: Deep Learning Accurately Models Complex Multiquark Particle Structures.

The study of multiquark states, composite particles bound together by the strong nuclear force, presents a significant challenge to contemporary physics due to the complexities arising from colour interactions and computational demands. Researchers Wei-Lin Wu from Peking University, Lu Meng from Ruhr-Universität Bochum, and Shi-Lin Zhu, also of Peking University, address these difficulties in their work, entitled ‘DeepQuark: deep-neural-network approach to multiquark bound states’. They present a novel application of deep neural networks to variational Monte Carlo calculations, a computational method used to find solutions to the Schrödinger equation, enabling the investigation of nucleon, tetraquark and pentaquark systems with improved efficiency and accuracy compared to existing techniques such as diffusion Monte Carlo and the Gaussian expansion method. Their approach demonstrates particular promise in modelling pentaquarks and exploring the underlying mechanisms of confinement within multiquark states, potentially offering new insights into non-perturbative quantum chromodynamics (QCD) and broader particle physics.
Multiquark systems pose a substantial computational challenge to theoretical physicists, necessitating approaches that exceed those typically employed for electron or nucleon systems. Researchers have recently introduced DeepQuark, a novel variational Monte Carlo (VMC) method utilising deep neural networks to address these intricate quantum many-body problems, with a specific focus on understanding the strong SU(3) colour interactions governing quark behaviour. This innovative approach constructs a trial wave function using a deep neural network, optimising its parameters within the VMC framework to approximate the true ground state of multiquark systems and unlock insights into the fundamental forces shaping matter.

The development of DeepQuark addresses critical limitations inherent in existing computational methods, which struggle to accurately model the increased correlations and discrete numbers characteristic of multiquark systems. Variational Monte Carlo is a class of Monte Carlo methods used to calculate the ground state energy of a many-body system. DeepQuark overcomes these challenges by employing a flexible neural network architecture capable of capturing complex many-body effects and adapting to the intricacies of the strong force. The strong force, also known as the colour force, is one of the four fundamental forces of nature and is responsible for binding quarks together within hadrons, such as protons and neutrons.

Results demonstrate DeepQuark achieves competitive accuracy when calculating the ground state energies of various quark systems, including mesons, baryons, and tetraquarks, when compared to established methods like diffusion Monte Carlo and the Gaussian expansion method. Notably, the method surpasses existing calculations for pentaquarks, exemplified by the triply heavy pentaquark, and successfully incorporates three-flux-tube confinement interactions for nucleons without increasing computational cost. This enhanced accuracy and efficiency stem from the neural network’s ability to handle stronger correlations and discrete numbers, allowing researchers to tackle previously intractable problems. Flux-tube confinement describes the linear potential energy between quarks as they are separated, analogous to a stretched rubber band.

The method’s performance comparisons demonstrate DeepQuark’s competitiveness with established techniques across studies of nucleons, doubly heavy tetraquarks, and fully heavy tetraquarks. Researchers rigorously tested the method’s accuracy and efficiency against benchmark calculations, confirming its ability to deliver reliable results even for complex multiquark systems.

The ability to extend to larger multiquark systems represents a significant advancement, overcoming computational barriers encountered in conventional methods. DeepQuark provides a powerful framework for investigating confining mechanisms beyond two-body interactions within multiquark states, potentially revealing new insights into the strong force. This exploration may yield valuable insights into non-perturbative quantum chromodynamics (QCD) and broader aspects of particle physics beyond the Standard Model. Quantum chromodynamics is the theory describing the strong interaction between quarks and gluons, the fundamental particles that make up hadrons.

In the pentaquark sector, the method predicts weakly bound molecular states and their bottom partners, analogous to the molecular structure observed in other particle systems. These findings suggest a potential experimental search for these pentaquarks in the D-wave channel, providing a testable prediction that can guide future experimental investigations. The D-wave channel refers to a specific angular momentum state of the pentaquark, influencing its decay properties and detectability.

Researchers validated the method’s ability to accurately predict the properties of both loosely bound molecular states and tightly bound compact tetraquarks, showcasing its adaptability to different types of multiquark configurations. This versatility is crucial for exploring the full spectrum of possible multiquark states and understanding the underlying principles governing their formation.

The development of DeepQuark marks a significant step forward in the field of hadronic physics, offering a powerful new tool for exploring the complex world of multiquark systems. Researchers anticipate that this method will play a crucial role in unraveling the mysteries of exotic hadronic matter and advancing our understanding of the strong force. Future work will focus on extending DeepQuark to even more complex systems and exploring its potential for discovering new physics beyond the Standard Model.