Understanding the ML black box with simple descriptors to predict cluster–adsorbate interaction energy

Agarwal, Sheena; Mehta, Shweta; Joshi, Kavita

doi:10.1039/d0nj00633e

Cited by 12 publications

(8 citation statements)

References 71 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…Likewise, there is a sizable amount of procedures to encode information in terms of descriptors. Investigations [494], [495], [496] as to which of these techniques hold advantages over others are crucial. In Ref.…”

Section: E Technical Aspectsmentioning

confidence: 99%

A Deep Dive into Machine Learning Density Functional Theory for Materials Science and Chemistry

Fiedler¹,

Shah²,

Bußmann³

et al. 2021

Preprint

View full text Add to dashboard Cite

With the growth of computational resources, the scope of electronic structure simulations has increased greatly. Artificial intelligence and robust data analysis hold the promise to accelerate large-scale simulations and their analysis to hitherto unattainable scales. Machine learning is a rapidly growing field for the processing of such complex datasets. It has recently gained traction in the domain of electronic structure simulations, where density functional theory takes the prominent role of the most widely used electronic structure method. Thus, DFT calculations represent one of the largest loads on academic high-performance computing systems across the world. Accelerating these with machine learning can reduce the resources required and enables simulations of larger systems. Hence, the combination of density functional theory and machine learning has the potential to rapidly advance electronic structure applications such as in-silico materials discovery and the search for new chemical reaction pathways. We provide the theoretical background of both density functional theory and machine learning on a generally accessible level. This serves as the basis of our comprehensive review including research articles up to December 2020 in chemistry and materials science that employ machine-learning techniques. In our analysis, we categorize the body of research into main threads and extract impactful results. We conclude our review with an outlook on exciting research directions in terms of a citation analysis.

show abstract

Section: E Technical Aspectsmentioning

confidence: 99%

A Deep Dive into Machine Learning Density Functional Theory for Materials Science and Chemistry

Fiedler¹,

Shah²,

Bußmann³

et al. 2021

Preprint

View full text Add to dashboard Cite

show abstract

“…Finally, their ML‐based model can predict the appropriate class of alloys useful for CO 2 reduction. Another advantage of tree‐based methods is that analysis of feature ranking is useful in understanding hidden trends in the dataset [49–52] . In a recent study, Liu et al.…”

Section: Introductionmentioning

confidence: 99%

Looking beyond Adsorption Energies to Understand Interactions at Surface using Machine Learning

Agarwal

Joshi

2022

ChemistrySelect

View full text Add to dashboard Cite

Identifying factors that influence interactions at the surface is still an active area of research. In this work, the importance of analyzing bond length activations (BLact) along with adsorption energies (Ea) while interpreting Density Functional Theory (DFT) results is emphasized. Investigating adsorption of different small molecules, such as O2, N2, CO, and CO2, on commonly studied facets ((100), (110), and (111)) of seven fcc transition metal surfaces (M=Ag, Au, Cu, Ir, Rh, Pt, and Pd) demonstrates the missing linear correlation between Ea and BLact. Further, tree based Machine Learning (ML) models reinforce the missing linear correlation between the two parameters and also highlight the importance of analyzing both to develop a better understanding of adsorption at surfaces. The best performing Random Forest models have a mean absolute error (MAE) of 0.19 eV for Ea prediction, and even lower MAE of 0.012 Å for BLact prediction. While often d‐band center is correlated with Ea, our observations show that infact the d‐band center has a better correlation with BLact. These observations emphasizes the role of BLact in gaining a fuller picture for catalysis. The fact that the factors responsible for BLact is a lesser‐explored subject adds to the novelty of the findings.

show abstract

“…[12,13,14,15,16,17,18] Machine Learning techniques are finding increasing application to problems of materials science, right from exploring suitable materials and structures for a desired property dependent application to digging out hidden patterns in the ML datasets. [19,20,21,22,23] The power of ML to assist domain experts with insights from vast datasets has proven to be of immense promise. [17,24,25,26] An important factor that lies at the heart of any ML problem is accurate "representation of the data".…”

Section: Introductionmentioning

confidence: 99%

DART: Deep Learning Enabled Topological Interaction Model for Energy Prediction of Metal Clusters and its Application in Identifying Unique Low Energy Isomers

Modee¹,

Agarwal²,

Verma³

et al. 2021

Preprint

Self Cite

View full text Add to dashboard Cite

<div><div><div><p>Recently, Machine Learning (ML) has proven to yield fast and accurate predictions of chemical properties to accelerate the discovery of novel molecules and materials. The majority of the work is on organic molecules, and much more work needs to be done for inorganic molecules, especially clusters. In the present work, we introduce a simple Topological Atomic Descriptor called TAD, which encodes chemical environment information of each atom in the cluster. TAD is a simple and interpretable descriptor where each value represents the atom count in three shells. We also introduce the DART, Deep Learning Enabled Topological Interaction model, which uses TAD as a feature vector to predict energies of metal clusters, in our case Gallium clusters with size ranging from 31 to 70 atoms. DART model is designed based on the principle that energy is a function of atomic interactions and allows us to model these complex atomic interactions to predict the energy. We further introduce a new dataset called GNC_31-70, which comprises structures and DFT optimized energies of Gallium clusters with sizes ranging from 31 to 70 atoms. We show how DART can be used to accelerate the identification of ground-state structures without geometry optimization. Albeit using topological descriptor, DART achieves MAE of 3.59 kcal/mol (0.15 eV) on testset. We also show that our model can distinguish core and surface atoms in the Ga-70 cluster, which the model has never encountered earlier. Finally, we demonstrate the transferability of DART model by predicting energies for about 6k unseen configurations picked up from Molecular Dynamics (MD) data for three cluster sizes (46, 57, and 60) within seconds. The DART model was able to reduce the load on DFT optimizations while identifying unique low energy structures from MD data.</p></div></div></div>

show abstract

Understanding the ML black box with simple descriptors to predict cluster–adsorbate interaction energy

Abstract: Density functional theory (DFT) is currently one of the most accurate and yet practical theories used to gain insight into the properties of materials.

Cited by 12 publications

References 71 publications

A Deep Dive into Machine Learning Density Functional Theory for Materials Science and Chemistry

A Deep Dive into Machine Learning Density Functional Theory for Materials Science and Chemistry

Looking beyond Adsorption Energies to Understand Interactions at Surface using Machine Learning

DART: Deep Learning Enabled Topological Interaction Model for Energy Prediction of Metal Clusters and its Application in Identifying Unique Low Energy Isomers

Contact Info

Product

Resources

About