Liu Xiaotong

XXB524, Changping

I completed my undergraduate and master’s degrees at the University of Science and Technology Beijing (USTB) in 2009 and 2012, respectively, and earned my Ph.D. from the University of Chinese Academy of Sciences (UCAS) in 2021. In January 2023, I was appointed as an Associate Professor at BISTU. I am also a member of Beijing Advanced Innovation Center for Materials Genome Engineering, BISTU. From February 2023 to February 2024, I served as a visiting scholar at Université catholique de Louvain (UCLouvain) in Belgium for one year (funded by CSC/WBI/FNRS Scholarship, five individuals from China each year). My academic journey was shaped by the guidance of two mentors, Prof. Hailei Zhao and Prof. Xiaodong Wen, along with a “virtual mentor” from UCLouvain, Prof. Gian-Marco Rignanese. Inspired by their footsteps, my current research primarily focuses on machine learning and theoretical computation within the field of materials science, often referred to by the trendy term “AI4Science”.

“All models are wrong, but some are useful”, as the saying goes. Different computational methods exhibit varying errors when calculating the same physical quantities; these discrepancies vary by material and computational approach. A key aspect of my research involves leveraging these error patterns in multi-fidelity data-driven models to more accurately predict the physical properties of actual materials. Another primary area of my focus is generative AI. I am currently endeavoring to develop improved descriptors and harness techniques like Variational Autoencoders (VAEs) and Large Language Model (LLM)-based generative models to guide material design. The descriptors I design are tested in the Matbench to evaluate their performance. Along these research trajectories, I encounter intriguing minor problems, which I abstract into challenges suitable for master’s and undergraduate students, offering a collaborative exploration of the field.

In addition to my academic pursuits, I have diverse professional experience having worked with a self-driving startup (UISEE), an internet company (Renren), and a large state-owned enterprise (PetroChina). These roles have broadened my perspective on problem-solving. I frequently share these experiences with my students and, if they show interest, I am open to collaborating with them on related projects. For instance, I collaborated with some students to develop a website for researchers, http://reaction.center/ .

news

Oct 25, 2023	A report from IJMMM.
Feb 20, 2023	A report from Prof. Siqi Shi
Nov 14, 2022	A report from BISTU news.

selected publications

A simple denoising approach to exploit multi-fidelity data for machine learning materials properties

Xiaotong Liu, Pierre-Paul De Breuck, Linghui Wang, and 1 more author

npj Computational Materials, 2022

Abs HTML

Machine-learning models have recently encountered enormous success for predicting the properties of materials. These are often trained based on data that present various levels of accuracy, with typically much less high- than low-fidelity data. In order to extract as much information as possible from all available data, we here introduce an approach which aims to improve the quality of the data through denoising. We investigate the possibilities that it offers in the case of the prediction of the band gap using both limited experimental data and density-functional theory relying on different exchange-correlation functionals. After analyzing the raw data thoroughly, we explore different ways to combine the data into training sequences and analyze the effect of the chosen denoiser. We also study the effect of applying the denoising procedure several times until convergence. Finally, we compare our approach with various existing methods to exploit multi-fidelity data and show that it provides an interesting improvement.
Solving chemistry problems via an end-to-end approach: A proof of concept

Xiaotong Liu, Tianfu Zhang, Tao Yang, and 7 more authors

The Journal of Physical Chemistry A, 2020

Abs HTML

Traditionally, chemistry problems are solved by means of a deductive approach. The question to be addressed is typically related to the value of a property that is either measured experimentally, computed using quantum-chemistry software, or (more recently) predicted using a machine-learned model. In this paper, we demonstrate that an inductive approach can be adopted using End-to-End (E2E) machine learning. This approach is illustrated for tackling the following chemistry problems: (i) determine the fully coordinated (FC) and undercoordinated (UC) atoms in a molecule with one missing atom, (ii) identify the type of atom that is missing in such an incomplete molecule, and (iii) predict the direction of a reaction between two molecules according to an existing dataset. The E2E approach leads to accuracies higher than 99%, 98%, and 93% for these three problems, respectively. Finally, in order to achieve such accuracies, a descriptor for the molecules, called bag of clusters, is introduced and compared with a series previously proposed descriptors, highlighting a series of advantages.
Lattice characteristics, structure stability and oxygen permeability of \(BaFe_{1-x}Y_xO_{3−δ}\) ceramic membranes

Xiaotong Liu, Hailei Zhao, Jianying Yang, and 5 more authors

Journal of membrane science, 2011

Abs HTML

\(BaFe_{1−x}Y_xO_{3−δ}\) (x=0–0.2) materials were synthesized by conventional solid-state reaction process for oxygen separation application. The effects of Y-doping on the crystal structure development, electrical conductivity and oxygen permeability were evaluated. Yttrium introduction effectively stabilize the cubic structure of BaFe1−xYxO3−δ. With Y-doping, the oxidation state of Fe ions reduces, resulting in the increase in oxygen vacancy concentration as charge compensation and the decreases in electrical conductivity. Y-doping enhances the structural stability of BaFe1−xYxO3−δ in reducing atmosphere but decreases the oxygen permeability. Both of them are attributed to the strong binding energy of Y–O bond. The cobalt free membrane BaFe0.95Y0.05O3−δ shows good structural stability under reducing atmosphere and acceptable oxygen permeation flux of 0.798 ml (STP) min−1 cm−2 at 900 °C for 1.1 mm thick membrane, making it a promising candidate for future practical applications.