SciPost Physics

SciPost Phys. 10, 073 (2021) · published 24 March 2021

• doi: 10.21468/SciPostPhys.10.3.073
• pdf
• BiBTeX
• RIS
• Submissions/Reports
• 

Abstract

We investigate the supervised machine learning of few interacting bosons in optical speckle disorder via artificial neural networks. The learning curve shows an approximately universal power-law scaling for different particle numbers and for different interaction strengths. We introduce a network architecture that can be trained and tested on heterogeneous datasets including different particle numbers. This network provides accurate predictions for all system sizes included in the training set and, by design, is suitable to attempt extrapolations to (computationally challenging) larger sizes. Notably, a novel transfer-learning strategy is implemented, whereby the learning of the larger systems is substantially accelerated and made consistently accurate by including in the training set many small-size instances.

• Tian et al., Machine learning for structure-property mapping of Ising models: Scalability and limitations
  Phys. Rev. E 108, 065304 (2023) [Crossref]
• Costa et al., Deep-learning density functionals for gradient descent optimization
  Phys. Rev. E 106, 045309 (2022) [Crossref]
• Kulik et al., Roadmap on Machine learning in electronic structure
  Electron. Struct. 4, 023004 (2022) [Crossref]
• Costa et al., Deep learning nonlocal and scalable energy functionals for quantum Ising models
  Phys. Rev. B 108, 125113 (2023) [Crossref]
• Mujal, Quantum Reservoir Computing for Speckle Disorder Potentials
  Condensed Matter 7, 17 (2022) [Crossref]
• Cantori et al., Supervised learning of random quantum circuits via scalable neural networks
  Quantum Sci. Technol. 8, 025022 (2023) [Crossref]
• Keeble et al., Machine learning one-dimensional spinless trapped fermionic systems with neural-network quantum states
  Phys. Rev. A 108, 063320 (2023) [Crossref]