The applications of artificial intelligence, machine learning, and deep learning techniques in the field of materials science are becoming increasingly common due to their promising abilities to extract and utilize data-driven information from available data and accelerate materials discovery and design for future applications. In an attempt to assist with this process, we deploy predictive models for multiple material properties, given the composition of the material. The deep learning models described here are built using a cross-property deep transfer learning technique, which leverages source models trained on large data sets to build target models on small data sets with different properties. We deploy these models in an online software tool that takes a number of material compositions as input, performs preprocessing to generate composition-based attributes for each material, and feeds them into the predictive models to obtain up to 41 different material property values. The material property predictor is available online at http://ai.eecs.northwestern.edu/MPpredictor.

Download full-text PDF

Source
http://www.ncbi.nlm.nih.gov/pmc/articles/PMC10091406PMC
http://dx.doi.org/10.1021/acs.jcim.3c00307DOI Listing

Publication Analysis

Top Keywords

material property
12
deep learning
8
predictive models
8
data sets
8
material
7
models
6
mppredictor artificial
4
artificial intelligence-driven
4
intelligence-driven web
4
web tool
4

Similar Publications

Reline, which is composed of choline chloride and urea in a molar ratio of 1:2, is the first and most extensively studied deep eutectic solvent (DES). In certain applications, reline is blended with organic solvents, dimethyl sulfoxide (DMSO) in most cases, to gain improved properties. Therefore, it is crucial to have a profound understanding of the impact of DMSO on the dynamics and structures of the species in the binary mixtures.

View Article and Find Full Text PDF

Exploring the Unique Properties and Superior Schwann Cell Guiding Abilities of Spider Egg Sac Silk.

ACS Appl Bio Mater

January 2025

Institute of Physics and Materials Science, Department of Natural Sciences and Sustainable Ressources, BOKU University, Peter Jordan-Straß 82, 1190 Vienna, Austria.

Spider silk (SPSI) is a promising candidate for use as a filler material in nerve guidance conduits (NGCs), facilitating peripheral nerve regeneration by providing a scaffold for Schwann cells (SCs) and axonal growth. However, the specific properties of SPSI that contribute to its regenerative success remain unclear. In this study, the egg sac silk of is investigated, which contains two distinct fiber types: tubuliform (TU) and major ampullate (MA) silk.

View Article and Find Full Text PDF

Ferrites with a Minimized Secondary Electron Yield.

Adv Sci (Weinh)

January 2025

School of Science and Engineering, University of Dundee, Nethergate, Dundee, DD1 4HN, UK.

Ferrites are an essential material in modern industry due to their exceptional magnetic properties and high resistivity. Many applications of ferrites necessitate exposure to high energy electrons, particularly space science and particle accelerators, where charging, multipacting, and electron clouds (ECs) are major issues. ECs are of particular concern around the Ni/Zn soft ferrite kicker magnets as the large hadron collider (LHC) undergoes its high luminosity upgrade.

View Article and Find Full Text PDF

A new [DyBiOCl(saph)] () Werner-type cluster has been prepared, which is the first Dy/Bi polynuclear compound with no metal-metal bond and one of the very few Ln-Bi (Ln = lanthanide) heterometallic complexes reported to date. The molecular compound has been deliberately transformed to its 1-D analogue [DyBiO(N)(saph)] () via the replacement of the terminal Cl ions by end-to-end bridging N groups. The overall metallic skeleton of (and ) can be described as consisting of a diamagnetic {Bi} unit with an elongated trigonal bipyramidal topology, surrounded by a magnetic {Dy} equilateral triangle, which does not contain μ-oxo/hydroxo/alkoxo groups.

View Article and Find Full Text PDF

We have successfully prepared a significant number of nanowires from non-toxic silicon sources. Compared to the SiO silicon source used in most other articles, our preparation method is much safer. It provides a simple and harmless new preparation method for the preparation of silicon nanowires.

View Article and Find Full Text PDF

Want AI Summaries of new PubMed Abstracts delivered to your In-box?

Enter search terms and have AI summaries delivered each week - change queries or unsubscribe any time!