Jiaxin Xu
Current Organizations / Institutes
The organizations/institutes from the scholar's latest article
Jiaxin Xu has a total of 25 co-authors. Here are the top co-authors by number of shared publications:
- Tengfei Luo
Department of Aerospace and Mechanical Engineering, University of Notre Dame, Notre Dame, USA
10 shared publications - Meng Jiang
University of Notre Dame
5 shared publications - Renzheng Zhang
University of Notre Dame
4 shared publications - Gang Liu
University of Notre Dame, Notre Dame, IN, USA
3 shared publications - Sobin Alosious
Lucy Family Institute for Data and Society, University of Notre Dame, Notre Dame, IN, 46556, USA
2 shared publications - Yuhan Liu
Department of Chemistry and Biochemistry, University of Notre Dame, Notre Dame, IN, 46556, USA
2 shared publications
Latest Publications
- Open Access
- 17 June 2026
- Published by Royal Society of Chemistry (RSC) in Digital Discovery
- Vol. 5 (7) , 3021-3043
ADEPT-PolyGraphMT: automated molecular simulation and multi-task multi-fidelity machine learning for polymer property generation and prediction
An integrated framework combining automated molecular simulations and multi-task, multi-fidelity machine learning for scalable polymer property generation, prediction, and large-scale screening.
- 17 June 2026
- Published by Royal Society of Chemistry (RSC) in Digital Discovery
- Vol. 5 (7) , 3021-3043
Open AccessADEPT-PolyGraphMT: automated molecular simulation and multi-task multi-fidelity machine learning for polymer property generation and prediction
An integrated framework combining automated molecular simulations and multi-task, multi-fidelity machine learning for scalable polymer property generation, prediction, and large-scale screening.
- Open Access
- 4 February 2026
- Published by Royal Society of Chemistry (RSC) in Polymer Chemistry
- Vol. 17 (10) , 1028-1051
A transfer learning framework integrating molecular dynamics and group contribution methods for predicting polymer specific heat capacity
A multi-fidelity transfer learning framework integrates group contribution, molecular dynamics, and experimental data to enable accurate and scalable prediction of polymer specific heat capacities.
- 4 February 2026
- Published by Royal Society of Chemistry (RSC) in Polymer Chemistry
- Vol. 17 (10) , 1028-1051
Open AccessA transfer learning framework integrating molecular dynamics and group contribution methods for predicting polymer specific heat capacity
A multi-fidelity transfer learning framework integrates group contribution, molecular dynamics, and experimental data to enable accurate and scalable prediction of polymer specific heat capacities.
- Open Access
- 6 February 2025
- Published by Royal Society of Chemistry (RSC) in Digital Discovery
- Vol. 4 (3) , 812-823
Active learning-guided exploration of thermally conductive polymers under strain
Active learning guided screening in combination with molecular dynamics simulation accelerate the discovery of high TC strained polymers and enhance the model performance in small data scenarios.
- 6 February 2025
- Published by Royal Society of Chemistry (RSC) in Digital Discovery
- Vol. 4 (3) , 812-823
Open AccessActive learning-guided exploration of thermally conductive polymers under strain
Active learning guided screening in combination with molecular dynamics simulation accelerate the discovery of high TC strained polymers and enhance the model performance in small data scenarios.
- 1 December 2024
- Published by AIP Publishing in Chemical Physics Reviews
- Vol. 5 (4)
Transcend the boundaries: Machine learning for designing polymeric membrane materials for gas separation
Polymeric membranes have become essential for energy-efficient gas separations such as natural gas sweetening, hydrogen separation, and carbon dioxide capture. Polymeric membranes face challenges like permeability-selectivity tradeoffs, plasticization, and physical aging, limiting their broader applicability. Machine learning (ML) techniques are increasingly used to address these challenges. This review covers current ML applications in polymeric gas separation membrane design, focusing on three key components: polymer data, representation methods, and ML algorithms. Exploring diverse polymer datasets related to gas separation, encompassing experimental, computational, and synthetic data, forms the foundation of ML applications. Various polymer representation methods are discussed, ranging from traditional descriptors and fingerprints to deep learning-based embeddings. Furthermore, we examine diverse ML algorithms applied to gas separation polymers. It provides insights into fundamental concepts such as supervised and unsupervised learning, emphasizing their applications in the context of polymer membranes. The review also extends to advanced ML techniques, including data-centric and model-centric methods, aimed at addressing challenges unique to polymer membranes, focusing on accurate screening and inverse design.
- 1 December 2024
- Published by AIP Publishing in Chemical Physics Reviews
- Vol. 5 (4)
Transcend the boundaries: Machine learning for designing polymeric membrane materials for gas separation
Polymeric membranes have become essential for energy-efficient gas separations such as natural gas sweetening, hydrogen separation, and carbon dioxide capture. Polymeric membranes face challenges like permeability-selectivity tradeoffs, plasticization, and physical aging, limiting their broader applicability. Machine learning (ML) techniques are increasingly used to address these challenges. This review covers current ML applications in polymeric gas separation membrane design, focusing on three key components: polymer data, representation methods, and ML algorithms. Exploring diverse polymer datasets related to gas separation, encompassing experimental, computational, and synthetic data, forms the foundation of ML applications. Various polymer representation methods are discussed, ranging from traditional descriptors and fingerprints to deep learning-based embeddings. Furthermore, we examine diverse ML algorithms applied to gas separation polymers. It provides insights into fundamental concepts such as supervised and unsupervised learning, emphasizing their applications in the context of polymer membranes. The review also extends to advanced ML techniques, including data-centric and model-centric methods, aimed at addressing challenges unique to polymer membranes, focusing on accurate screening and inverse design.
- Open Access
- 3 September 2024
- Published by IOP Publishing in Progress in Energy
- Vol. 6 (4) , 042005
A review on machine learning-guided design of energy materials
The development and design of energy materials are essential for improving the efficiency, sustainability, and durability of energy systems to address climate change issues. However, optimizing and developing energy materials can be challenging due to large and complex search spaces. With the advancements in computational power and algorithms over the past decade, machine learning (ML) techniques are being widely applied in various industrial and research areas for different purposes. The energy material community has increasingly leveraged ML to accelerate property predictions and design processes. This article aims to provide a comprehensive review of research in different energy material fields that employ ML techniques. It begins with foundational concepts and a broad overview of ML applications in energy material research, followed by examples of successful ML applications in energy material design. We also discuss the current challenges of ML in energy material design and our perspectives. Our viewpoint is that ML will be an integral component of energy materials research, but data scarcity, lack of tailored ML algorithms, and challenges in experimentally realizing ML-predicted candidates are major barriers that still need to be overcome.
- 3 September 2024
- Published by IOP Publishing in Progress in Energy
- Vol. 6 (4) , 042005
Open AccessA review on machine learning-guided design of energy materials
The development and design of energy materials are essential for improving the efficiency, sustainability, and durability of energy systems to address climate change issues. However, optimizing and developing energy materials can be challenging due to large and complex search spaces. With the advancements in computational power and algorithms over the past decade, machine learning (ML) techniques are being widely applied in various industrial and research areas for different purposes. The energy material community has increasingly leveraged ML to accelerate property predictions and design processes. This article aims to provide a comprehensive review of research in different energy material fields that employ ML techniques. It begins with foundational concepts and a broad overview of ML applications in energy material research, followed by examples of successful ML applications in energy material design. We also discuss the current challenges of ML in energy material design and our perspectives. Our viewpoint is that ML will be an integral component of energy materials research, but data scarcity, lack of tailored ML algorithms, and challenges in experimentally realizing ML-predicted candidates are major barriers that still need to be overcome.
- Open Access
- 26 June 2024
- Published by Elsevier in Cell Reports Physical Science
- Vol. 5 (7)
Superior polymeric gas separation membrane designed by explainable graph machine learning
- 26 June 2024
- Published by Elsevier in Cell Reports Physical Science
- Vol. 5 (7)
Open AccessSuperior polymeric gas separation membrane designed by explainable graph machine learning
- Open Access
- 16 April 2024
- Published by Springer Nature in npj Computational Materials
- Vol. 10 (1) , 1-12
Unlocking enhanced thermal conductivity in polymer blends through active learning
Polymers play an integral role in various applications, from everyday use to advanced technologies. In the era of machine learning (ML), polymer informatics has become a vital field for efficiently designing and developing polymeric materials. However, the focus of polymer informatics has predominantly centered on single-component polymers, leaving the vast chemical space of polymer blends relatively unexplored. This study employs a high-throughput molecular dynamics (MD) simulation combined with active learning (AL) to uncover polymer blends with enhanced thermal conductivity (TC) compared to the constituent single-component polymers. Initially, the TC of about 600 amorphous single-component polymers and 200 amorphous polymer blends with varying blending ratios are determined through MD simulations. The optimal representation method for polymer blends is identified, which involves a weighted sum approach that extends existing polymer representation from single-component polymers to polymer blends. An AL framework, combining MD simulation and ML, is employed to explore the TC of approximately 550,000 unlabeled polymer blends. The AL framework proves highly effective in accelerating the discovery of high-performance polymer blends for thermal transport. Additionally, we delve into the relationship between TC, radius of gyration (Rg), and hydrogen bonding, highlighting the roles of inter- and intra-chain interactions in thermal transport in amorphous polymer blends. A significant positive association between TC and Rg improvement and an indirect contribution from H-bond interaction to TC enhancement are revealed through a log-linear model and an odds ratio calculation, emphasizing the impact of increasing Rg and H-bond interactions on enhancing polymer blend TC.
- 16 April 2024
- Published by Springer Nature in npj Computational Materials
- Vol. 10 (1) , 1-12
Open AccessUnlocking enhanced thermal conductivity in polymer blends through active learning
Polymers play an integral role in various applications, from everyday use to advanced technologies. In the era of machine learning (ML), polymer informatics has become a vital field for efficiently designing and developing polymeric materials. However, the focus of polymer informatics has predominantly centered on single-component polymers, leaving the vast chemical space of polymer blends relatively unexplored. This study employs a high-throughput molecular dynamics (MD) simulation combined with active learning (AL) to uncover polymer blends with enhanced thermal conductivity (TC) compared to the constituent single-component polymers. Initially, the TC of about 600 amorphous single-component polymers and 200 amorphous polymer blends with varying blending ratios are determined through MD simulations. The optimal representation method for polymer blends is identified, which involves a weighted sum approach that extends existing polymer representation from single-component polymers to polymer blends. An AL framework, combining MD simulation and ML, is employed to explore the TC of approximately 550,000 unlabeled polymer blends. The AL framework proves highly effective in accelerating the discovery of high-performance polymer blends for thermal transport. Additionally, we delve into the relationship between TC, radius of gyration (Rg), and hydrogen bonding, highlighting the roles of inter- and intra-chain interactions in thermal transport in amorphous polymer blends. A significant positive association between TC and Rg improvement and an indirect contribution from H-bond interaction to TC enhancement are revealed through a log-linear model and an odds ratio calculation, emphasizing the impact of increasing Rg and H-bond interactions on enhancing polymer blend TC.
- Open Access
- 28 December 2023
- Published by Springer Nature in Research Square
Unlocking Enhanced Thermal Conductivity in Polymer Blends through Active Learning
Polymers play an integral role in various applications, from everyday use to advanced technologies. In the era of machine learning (ML), polymer informatics has emerged as a vital field for efficiently designing and developing polymeric materials. However, the focus of polymer informatics has predominantly centered on single-component polymers, leaving the vast chemical space of polymer blends relatively unexplored. This study employs a high-throughput molecular dynamics (MD) simulation approach combined with active learning (AL) to uncover polymer blends with enhanced thermal conductivity (TC) compared to the constituent single-component polymers. Initially, TC values for about 600 amorphous single-component polymers and approximately 200 amorphous polymer blends with varying blending ratios are determined through MD simulations. The optimal representation method for polymer blends is also identified, which involves a weighted sum approach that extends existing polymer representation from single-component polymers to polymer blends. An AL framework, combining MD simulation and ML models, is employed to explore the TC of an unlabeled dataset comprising approximately 550,000 potential polymer blends. The AL framework proves highly effective in accelerating the discovery of high-performance polymer blends for thermal transport. Additionally, we delve into the relationship between TC, radius of gyration (Rg), and hydrogen bonding, highlighting the roles of inter- and intra-chain interactions in thermal transport in amorphous polymer blends. A significant positive association between TC and Rg improvement and an indirect contribution from H-bond interaction to TC enhancement are revealed through a log-linear model and odd ratio calculation based on the polymer blend dataset, emphasizing the impact of increasing Rg and H-bond interactions on enhancing polymer blend TC.
- 28 December 2023
- Published by Springer Nature in Research Square
Open AccessUnlocking Enhanced Thermal Conductivity in Polymer Blends through Active Learning
Polymers play an integral role in various applications, from everyday use to advanced technologies. In the era of machine learning (ML), polymer informatics has emerged as a vital field for efficiently designing and developing polymeric materials. However, the focus of polymer informatics has predominantly centered on single-component polymers, leaving the vast chemical space of polymer blends relatively unexplored. This study employs a high-throughput molecular dynamics (MD) simulation approach combined with active learning (AL) to uncover polymer blends with enhanced thermal conductivity (TC) compared to the constituent single-component polymers. Initially, TC values for about 600 amorphous single-component polymers and approximately 200 amorphous polymer blends with varying blending ratios are determined through MD simulations. The optimal representation method for polymer blends is also identified, which involves a weighted sum approach that extends existing polymer representation from single-component polymers to polymer blends. An AL framework, combining MD simulation and ML models, is employed to explore the TC of an unlabeled dataset comprising approximately 550,000 potential polymer blends. The AL framework proves highly effective in accelerating the discovery of high-performance polymer blends for thermal transport. Additionally, we delve into the relationship between TC, radius of gyration (Rg), and hydrogen bonding, highlighting the roles of inter- and intra-chain interactions in thermal transport in amorphous polymer blends. A significant positive association between TC and Rg improvement and an indirect contribution from H-bond interaction to TC enhancement are revealed through a log-linear model and odd ratio calculation based on the polymer blend dataset, emphasizing the impact of increasing Rg and H-bond interactions on enhancing polymer blend TC.
- Open Access
Data-Centric Learning from Unlabeled Graphs with Diffusion Model
Graph property prediction tasks are important and numerous. While each task offers a small size of labeled examples, unlabeled graphs have been collected from various sources and at a large scale. A conventional approach is training a model with the unlabeled graphs on self-supervised tasks and then fine-tuning the model on the prediction tasks. However, the self-supervised task knowledge could not be aligned or sometimes conflicted with what the predictions needed. In this paper, we propose to extract the knowledge underlying the large set of unlabeled graphs as a specific set of useful data points to augment each property prediction model. We use a diffusion model to fully utilize the unlabeled graphs and design two new objectives to guide the model's denoising process with each task's labeled data to generate task-specific graph examples and their labels. Experiments demonstrate that our data-centric approach performs significantly better than fifteen existing various methods on fifteen tasks. The performance improvement brought by unlabeled data is visible as the generated labeled examples unlike the self-supervised learning.
Open AccessData-Centric Learning from Unlabeled Graphs with Diffusion Model
Graph property prediction tasks are important and numerous. While each task offers a small size of labeled examples, unlabeled graphs have been collected from various sources and at a large scale. A conventional approach is training a model with the unlabeled graphs on self-supervised tasks and then fine-tuning the model on the prediction tasks. However, the self-supervised task knowledge could not be aligned or sometimes conflicted with what the predictions needed. In this paper, we propose to extract the knowledge underlying the large set of unlabeled graphs as a specific set of useful data points to augment each property prediction model. We use a diffusion model to fully utilize the unlabeled graphs and design two new objectives to guide the model's denoising process with each task's labeled data to generate task-specific graph examples and their labels. Experiments demonstrate that our data-centric approach performs significantly better than fifteen existing various methods on fifteen tasks. The performance improvement brought by unlabeled data is visible as the generated labeled examples unlike the self-supervised learning.