Article Contents
Article Contents

# Analytic continuation of noisy data using Adams Bashforth residual neural network

• * Corresponding author
• We propose a data-driven learning framework for the analytic continuation problem in numerical quantum many-body physics. Designing an accurate and efficient framework for the analytic continuation of imaginary time using computational data is a grand challenge that has hindered meaningful links with experimental data. The standard Maximum Entropy (MaxEnt)-based method is limited by the quality of the computational data and the availability of prior information. Also, the MaxEnt is not able to solve the inversion problem under high level of noise in the data. Here we introduce a novel learning model for the analytic continuation problem using a Adams-Bashforth residual neural network (AB-ResNet). The advantage of this deep learning network is that it is model independent and, therefore, does not require prior information concerning the quantity of interest given by the spectral function. More importantly, the ResNet-based model achieves higher accuracy than MaxEnt for data with higher level of noise. Finally, numerical examples show that the developed AB-ResNet is able to recover the spectral function with accuracy comparable to MaxEnt where the noise level is relatively small.

Mathematics Subject Classification: Primary: 45B05; Secondary: 32W50, 49N30.

 Citation:

• Figure 1.  Illustration of data-driven learning framework for analytic continuation

Figure 2.  Single hidden layer neural network structure

Figure 3.  Residual neural network block

Figure 4.  Multistep neural network architecture

Figure 5.  One data sample from the training set $G(\tau)$ (top left), Legendre representation $G_l$ (top right), and target spectral density $A(\omega)$ (bottom)

Figure 6.  The training performance from AB1-ResNet, AB2-ResNet, and AB3-ResNet structure with data noise $10^{-2}$

Figure 7.  Three different spectral density function $A(\omega)$ generated from AB3-ResNet and Maxent (dark line). The left column represents results from dataset with noise level $10^{-2}$, the right column shows results obtained from the dataset under noise level $10^{-3}$

Figure 8.  The comparison of predicted spectral function between different AB-ResNet

Figures(8)