Reconstructing Functions and Estimating Parameters with Artificial Neural Network: a test with Hubble parameter and SNe Ia. (arXiv:1910.03636v1 [astro-ph.CO])
<a href="http://arxiv.org/find/astro-ph/1/au:+Wang_G/0/1/0/all/0/1">Guo-Jian Wang</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Ma_X/0/1/0/all/0/1">Xiao-Jiao Ma</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Li_S/0/1/0/all/0/1">Si-Yao Li</a>, <a href="http://arxiv.org/find/astro-ph/1/au:+Xia_J/0/1/0/all/0/1">Jun-Qing Xia</a>

In this work, we propose a new non-parametric approach for reconstructing a
function from observational data using Artificial Neural Network (ANN), which
has no assumptions to the data and is a completely data-driven approach. We
test the ANN method by reconstructing functions of the Hubble parameter
measurements $H(z)$ and the distance redshift relation $D_L(z)$ of type Ia
supernova. We find that both $H(z)$ and $D_L(z)$ can be reconstructed with high
accuracy. Furthermore, we estimate cosmological parameters using the
reconstructed functions of $H(z)$ and $D_L(z)$ and find the results are
consistent with those obtained using the observational data directly.
Therefore, we propose that the function reconstructed by ANN can represent the
actual distribution of observational data and can be used for parameter
estimation in further cosmological research. In addition, we present a new
strategy to train and evaluate the neural network, and a code for
reconstructing functions using ANN has been developed and will be available
soon.

In this work, we propose a new non-parametric approach for reconstructing a
function from observational data using Artificial Neural Network (ANN), which
has no assumptions to the data and is a completely data-driven approach. We
test the ANN method by reconstructing functions of the Hubble parameter
measurements $H(z)$ and the distance redshift relation $D_L(z)$ of type Ia
supernova. We find that both $H(z)$ and $D_L(z)$ can be reconstructed with high
accuracy. Furthermore, we estimate cosmological parameters using the
reconstructed functions of $H(z)$ and $D_L(z)$ and find the results are
consistent with those obtained using the observational data directly.
Therefore, we propose that the function reconstructed by ANN can represent the
actual distribution of observational data and can be used for parameter
estimation in further cosmological research. In addition, we present a new
strategy to train and evaluate the neural network, and a code for
reconstructing functions using ANN has been developed and will be available
soon.

http://arxiv.org/icons/sfx.gif