DART: deep learning enabled topological interaction model for energy prediction of metal clusters and its application in identifying unique low energy isomers

Modee, Rohit; Agarwal, Sheena; Verma, Ashwini; Joshi, Kavita

doi:10.1039/d1cp02956h

Cited by 10 publications

(12 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In our previous work, DART is trained over GNC_31-70 data-set to predict energy for a give Ga cluster. The GNC_31-70 comprised of low energy isomers and their corresponding energies with sizes ranging from 31 to 70 atoms [36]. In this work, we added new optimized geometries of neutral Ga n clusters (size n = 2-31) to data-set GNC_31-70.…”

Section: Datasetmentioning

confidence: 99%

“…As shown in figure 3, we generate gallium clusters by adding a single atom from the bag B to the existing seed structure of the gallium cluster present on the 3D canvas C. In principle, we are generating N atoms cluster from N − 1 atoms seed cluster already present on canvas C. The idea is to allow the model to learn the process of addition of a single atom to N − 1 atom cluster to generate N atom cluster. The reward function R a (s, s ′ ) is given in equation ( 4) as the negative of the energy calculated using DART model [36]. This allows the model to learn the most efficient ways of adding atoms in the presence of different structural motifs across different sizes/classes of gallium clusters to get low-energy gallium clusters.…”

Section: Rl Formulationmentioning

confidence: 99%

“…The reward function is based on a fundamental physical property (energy). We use the previously developed machine learning (ML) model DART [36] to calculate the energies of various conformers/isomers generated during training. Contrary to other models that use structural information, our DART model uses features that capture topological/connectivity information to predict metallic clusters energies.…”

Section: Introductionmentioning

confidence: 99%

See 2 more Smart Citations

MeGen - generation of gallium metal clusters using reinforcement learning

Modee

Verma

Joshi

2023

Mach. Learn.: Sci. Technol.

Self Cite

View full text Add to dashboard Cite

The generation of low-energy 3D structures of metal clusters depends on the efficiency of the search algorithm and the accuracy of inter-atomic interaction description. In this work, we formulate the search algorithm as a Reinforcement Learning (RL) problem. Concisely, we propose a novel actor-critic architecture that generates low-lying isomers of metal clusters at a fraction of computational cost than conventional methods. Our RL-based search algorithm uses a previously developed DART model as a reward function to describe the inter-atomic interactions to validate predicted structures. Using the DART model as a reward function incentivizes the RL model to generate low-energy structures and helps generate valid structures. We demonstrate the advantages of our approach over conventional methods for scanning local minima on potential energy surface (PES). Our approach not only generates isomer of gallium clusters at a minimal computational cost but also predicts isomer families that were not discovered through previous DFT-based approaches.

show abstract

Section: Datasetmentioning

confidence: 99%

Section: Rl Formulationmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 1 more Smart Citation

MeGen - generation of gallium metal clusters using reinforcement learning

Modee

Verma

Joshi

2023

Mach. Learn.: Sci. Technol.

Self Cite

View full text Add to dashboard Cite

show abstract

“…Neural network potentials (NNPs) learn to approximate the potential energy surface (PES) as a high dimensional function (HDF) f by learning from existing reference data. Once trained NNPs can successfully circumvent the need to solve the electronic Schrödinger equation explicitly as it has learned the mapping f ( Z i , r i ) → E , where Z i are the nuclear charges and r i are the atomic positions. − Machine learning (ML) methods in general have been successful in improving computational chemistry algorithms leading to accelerated property prediction and chemical space exploration . Recently, much emphasis has been on developing efficient ML-based search algorithms to explore chemical space, − but the same is not the case for conformational space.…”

Section: Introductionmentioning

confidence: 99%

MolOpt: Autonomous Molecular Geometry Optimization Using Multiagent Reinforcement Learning

Modee,

Mehta,

Laghuvarapu

et al. 2023

J. Phys. Chem. B

Self Cite

View full text Add to dashboard Cite

“…Other descriptors use topological information , or a graph-based representation, − both of which have deep roots in cheminformatics and find applications in machine learning for chemistry. Working a set of empirical topological descriptors into autocorrelation functions has been applied to transition metal complexes by Janet and Kulik .…”

Section: Introductionmentioning

confidence: 99%

Quantum Chemical Roots of Machine-Learning Molecular Similarity Descriptors

Gugler

Reiher

2022

J. Chem. Theory Comput.

View full text Add to dashboard Cite

In this work, we explore the quantum chemical foundations of descriptors for molecular similarity. Such descriptors are key for traversing chemical compound space with machine learning. Our focus is on the Coulomb matrix and on the smooth overlap of atomic positions (SOAP). We adopt a basic framework that allows us to connect both descriptors to electronic structure theory. This framework enables us to then define two new descriptors that are more closely related to electronic structure theory, which we call Coulomb lists and smooth overlap of electron densities (SOED). By investigating their usefulness as molecular similarity descriptors, we gain new insights into how and why Coulomb matrix and SOAP work. Moreover, Coulomb lists avoid the somewhat mysterious diagonalization step of the Coulomb matrix and might provide a direct means to extract subsystem information that can be compared across Born−Oppenheimer surfaces of varying dimension. For the electron density, we derive the necessary formalism to create the SOED measure in close analogy to SOAP. Because this formalism is more involved than that of SOAP, we review the essential theory as well as introduce a set of approximations that eventually allow us to work with SOED in terms of the same implementation available for the evaluation of SOAP. We focus our analysis on elementary reaction steps, where transition state structures are more similar to either reactant or product structures than the latter two are with respect to one another. The prediction of electronic energies of transition state structures can, however, be more difficult than that of stable intermediates due to multi-configurational effects. The question arises to what extent molecular similarity descriptors rooted in electronic structure theory can resolve these intricate effects.

show abstract

DART: deep learning enabled topological interaction model for energy prediction of metal clusters and its application in identifying unique low energy isomers

Abstract: 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...

Cited by 10 publications

References 58 publications

MeGen - generation of gallium metal clusters using reinforcement learning

MeGen - generation of gallium metal clusters using reinforcement learning

MolOpt: Autonomous Molecular Geometry Optimization Using Multiagent Reinforcement Learning

Quantum Chemical Roots of Machine-Learning Molecular Similarity Descriptors

Contact Info

Product

Resources

About