End-to-end material thermal conductivity prediction through machine learning

Yagyank Srivastava, Ankit Jain

Journal of Applied Physics 2023 article cited by 21

Source Open-access PDF More in Engineering & Materials

importance 4/10 partially solved Machine Learning in Materials Science materials science stage: research

Problem

Predicting the thermal conductivity of materials is slow and limited by scarce high-quality data.

Why it matters

Faster thermal conductivity prediction could speed development of heat-management materials.

What the AI did

After expanding the dataset with first-principles calculations for 225 materials, the authors assessed state-of-the-art ML models (all overfit) and introduced a graph-based neural network with more consistent, regularized performance.

Results

Best mean absolute percentage error of 50–60% on the test dataset; dataset of 225 materials more than doubled the existing one.

Impact

Models are useful for expediting material screening but too inaccurate for reliable prediction.

Record

Venue	Journal of Applied Physics
Published	2023
Work type	article
Authors	Yagyank Srivastava, Ankit Jain
Citations	21 (OpenAlex, at import time)
References	51
Model or method	graph-based neural network
Organization	not stated in source
Primary source	https://doi.org/10.1063/5.0183513
Open access	green
Keywords	thermal conductivity, machine learning, graph neural network, overfitting, materials screening

Importance rating

4 / 10

real but narrower advance

Faster thermal conductivity prediction could speed heat-management material screening, but abstract gives no scale of impact and model accuracy remains limited (50-60% MAPE).

Basis: inferred. Judged from the source text, which gives no explicit scale. Stated scale: not stated in source. Method and limits.

More in Engineering & Materials

#	Paper	Year	Citations	Importance
1	Scaling deep learning for materials discovery Amil Merchant, Simon Batzner, Samuel S. Schoenholz, Muratahan Aykol, G...	2023	1,471	8
2	Application of Machine Learning in Material Synthesis and Property Prediction Guannan Huang, Yani Guo, Ye Chen, Zhengwei Nie	2023	165	7
3	Machine learning-guided discovery of ionic polymer electrolytes for lithium metal batteries Kai Li, Jifeng Wang, Yuanyuan Song, Ying Wang	2023	115	7
4	Machine Learning for Structural Health Monitoring of Aerospace Structures: A Review Gennaro Scarselli, Francesco Nicassio	2025	44	7
5	Power System Transient Stability Assessment Based on Machine Learning Algorithms and Grid Topology Mihail D. Senyuk, Murodbek Safaraliev, Firuz Kamalov, Hana Sulieman	2023	42	7