# Improving neural network predictions of material properties with limited data using transfer learning

arXiv: https://arxiv.org/abs/2006.16420 · HTML: https://mlsanity.com/paper/2006.16420/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- First posted on arXiv 29 Jun 2020.

## Built on by (most cited first, 1 of 1)

- [Tackling Data Scarcity with Transfer Learning: A Case Study of Thickness Characterization from Optical Spectra of Perovskite Thin Films](https://mlsanity.com/paper/2207.02209/) (2022)

## Abstract (arXiv)

We develop new transfer learning algorithms to accelerate prediction of material properties from ab initio simulations based on density functional theory (DFT). Transfer learning has been successfully utilized for data-efficient modeling in applications other than materials science, and it allows transferable representations learned from large datasets to be repurposed for learning new tasks even with small datasets. In the context of materials science, this opens the possibility to develop generalizable neural network models that can be repurposed on other materials, without the need of generating a large (computationally expensive) training set of materials properties. The proposed transfer learning algorithms are demonstrated on predicting the Gibbs free energy of light transition metal oxides.