Spectral Reconstruction with Deep Neural Networks

Abstract

We explore artificial neural networks as a tool for the reconstruction of spectral functions from imaginary time Green's functions, a classic ill-conditioned inverse problem. Our ansatz is based on a supervised learning framework in which prior knowledge is encoded in the training data and the inverse transformation manifold is explicitly parametrised through a neural network. We systematically investigate this novel reconstruction approach, providing a detailed analysis of its performance on physically motivated mock data, and compare it to established methods of Bayesian inference. The reconstruction accuracy is found to be at least comparable, and potentially superior in particular at larger noise levels. We argue that the use of labelled training data in a supervised setting and the freedom in defining an optimisation objective are inherent advantages of the present approach and may lead to significant improvements over state-of-the-art methods in the future. Potential directions for further research are discussed in detail.

arXiv comment20 pages, 16 figures

References in corpus (22)

• Deep Learning in Neural Networks: An Overview
• Solving the Quantum Many-Body Problem with Artificial Neural Networks
• Machine learning phases of matter
• Discovering Phase Transitions with Unsupervised Learning
• Maximum Entropy Analysis of the Spectral Functions in Lattice QCD
• Unsupervised learning of phase transitions: from principal component analysis to variational autoencoders
• Deep Learning and its Application to LHC Physics
• Discovering Phases, Phase Transitions and Crossovers through Unsupervised Machine Learning: A critical examination
• Self-Learning Monte Carlo Method
• Accelerate Monte Carlo Simulations with Restricted Boltzmann Machines
• A novel Bayesian approach to spectral function reconstruction
• Machine Learning of Explicit Order Parameters: From the Ising Model to SU(2) Lattice Gauge Theory
• Reconstructing the gluon
• Reconstructing parton distribution functions from Ioffe time data: from Bayesian methods to Neural Networks
• Artificial Neural Network Approach to the Analytic Continuation Problem
• Learning to Teach with Dynamic Loss Functions
• Reconstructing the thermal Green functions at real times from those at imaginary times
• Analytic continuation via domain-knowledge free machine learning
• Machine learning action parameters in lattice quantum chromodynamics
• Towards Novel Insights in Lattice Field Theory with Explainable Machine Learning
• A test on analytic continuation of thermal imaginary-time data
• Reducing Autocorrelation Times in Lattice Simulations with Generative Adversarial Networks

Cited by in corpus (26)

• Exploring QCD matter in extreme conditions with Machine Learning
• Reconstructing QCD Spectral Functions with Gaussian Processes
• Lattice gauge equivariant convolutional neural networks
• Heavy quark potential in quark-gluon Plasma: Deep neural network meets lattice quantum chromodynamics
• Neural network reconstruction of the dense matter equation of state from neutron star observables
• Reconstructing the neutron star equation of state from observational data via automatic differentiation
• Electrical conductivity of the quark-gluon plasma: perspective from lattice QCD
• Reconstructing spectral functions via automatic differentiation
• Rethinking the ill-posedness of the spectral function reconstruction -- why is it fundamentally hard and how Artificial Neural Networks can help
• Spectral density reconstruction with Chebyshev polynomials
• Non-perturbative strong coupling at timelike momenta
• The QCD moat regime and its real-time properties
• Robust analytic continuation combining the advantages of the sparse modeling approach and Padé approximation
• Inverse problems, real-time dynamics and lattice simulations
• Bayesian inference of real-time dynamics from lattice QCD
• Neural network approach to reconstructing spectral functions and complex poles of confined particles
• Bryan's Maximum Entropy Method -- diagnosis of a flawed argument and its remedy
• Machine learning spectral functions in lattice QCD
• Reconstructing the graviton
• Towards constraining QCD phase transitions in neutron star interiors: Bayesian Inference with TOV linear response analysis
• Faster spectral density calculation using energy moments
• Reconstructing lattice QCD spectral functions with stochastic pole expansion and Nevanlinna analytic continuation
• Report from the A.I. For Nuclear Physics Workshop
• Application of radial basis functions neutral networks in spectral functions
• A Parameter-Free Differential Evolution Algorithm for the Analytic Continuation of Imaginary Time Correlation Functions
• Nuclear two point correlation functions on a quantum-computer