arXiv subjects

Arghya Bhowmik

Publications and source records attributed to Arghya Bhowmik.

At least 19 recordsLinked to original sources

arXiv2026

Complete Neural Electronic Initialization Accelerates Materials DFT

We present the first complete machine learning method for accelerating plane-wave density functional theory (DFT) in materials under the projector augmented wave (PAW) formalism. We formalize seven criteria that a \textit{Complete Neural Electronic Initializer} must satisfy for practical end-to-end PAW DFT acceleration. Applying these criteria to prior work reveals two missing structure-dependent components, augmentation occupancies and spin initialization, that prevent existing methods from providing complete reference-free initialization. Controlled ablations show that omitting these components can eliminate or reverse the acceleration obtained via models that only predict the smooth valence density. We satisfy these missing requirements by introducing AugNet, the first general equivariant model for PAW augmentation occupancies, and the first general spin density model for materials, which predicts the smooth spin-difference density and spin-difference PAW augmentation occupancies using predicted magnetic moments to constrain the global magnetic state. Combined with existing valence density models, these components satisfy all seven criteria and form a fully reference-free electronic initializer for materials DFT, requiring no electronic quantities from a converged target calculation. Our method reduces end-to-end DFT wall time by up to ~25% on unseen structures while preserving converged energies.

cond-mat.mtrl-sci↗

arXiv2026

HIP: Hessian Interatomic Potentials without derivatives

Molecular Hessians, the second derivatives of the potential energy, are fundamental to many workflows in computational chemistry. Usually, accurate Hessians are computationally expensive to calculate and scale poorly with system size, whether computed using quantum chemistry methods or machine-learning interatomic potentials (MLIPs). In this work, we introduce Hessian interatomic potentials (HIPs), a deep learning model that directly predicts Hessians without relying on automatic differentiation or finite differences. To do so, we construct SE(3)-equivariant, symmetric Hessians from irreducible representation (irrep) features up to degree $l$=2, computed by a graph neural network. HIP Hessians are one to two orders of magnitude faster, more accurate, more memory efficient, easier to train, and exhibit more favourable scaling with system size. We validate our predictions across a wide range of downstream tasks, demonstrating consistently superior performance in transition state search, geometry optimization, zero-point energy corrections, and vibrational analysis. We open-source the HIP code and model weights.

cs.LG↗

arXiv2026

Global Plane Waves From Local Gaussians: Periodic Charge Densities in a Blink

We introduce ELECTRAFI, a fast, end-to-end differentiable model for predicting periodic charge densities in crystalline materials. ELECTRAFI constructs anisotropic Gaussians in real space and exploits their closed-form Fourier transforms to analytically evaluate plane-wave coefficients via the Poisson summation formula. This formulation delegates non-local and periodic behavior to analytic transforms, enabling reconstruction of the full periodic charge density with a single inverse FFT. By avoiding explicit real-space grid probing, periodic image summation, and spherical harmonic expansions, ELECTRAFI matches or exceeds state-of-the-art accuracy across periodic benchmarks while being up to $633 \times$ faster than the strongest competing method, reconstructing crystal charge densities in a fraction of a second. When used to initialize DFT calculations, ELECTRAFI reduces total DFT compute cost by up to ~20%, whereas slower charge density models negate savings due to high inference times. Our results show that accuracy and inference cost jointly determine end-to-end DFT speedups, and motivate our focus on efficiency.

cond-mat.mtrl-sci↗

arXiv2026

AtomComposer: Discovering Chemical Space from First Principles with Reinforcement Learning

Discovering novel stable molecules without training data remains a grand scientific challenge. Current molecular generative models are trained on large, pre-curated datasets, which introduce biases and limit exploration of novel chemistry. In contrast, we propose a new paradigm: autonomous, generalized agents capable of mapping vast, unknown chemical spaces without any pretraining. For the first time, we present AtomComposer, a self-guided agent that autonomously constructs valid 3D isomers under stoichiometric constraints and is trained exclusively online using reinforcement learning. Unlike existing approaches that generally overfit to a specific chemical formula, we establish a multi-composition training scheme that enables a broad generalization across diverse chemistry, guided by energy- and validity-based rewards. Our agent can discover up to an order of magnitude more valid isomers on unseen test formulas than existing single-composition reinforcement-learning baselines trained with per-step energy rewards. These results fulfill the promise of online reinforcement learning as a powerful paradigm for scalable, from-scratch exploration of chemical configuration space.

cs.LG↗

arXiv2026

Designing dislocation-driven polar vortex networks in twisted perovskites

Twisting two atomic layers produces a geometric moire pattern, but bonding-induced interfacial reconstruction fundamentally transforms this into an ordered dislocation network - a distinction obscured in weakly-bonded van der Waals systems. Although in-plane topological vortex nanostructures arising from twisting-induced lateral strain modulation have been linked to periodic moire patterns in freestanding perovskite layers and 2D bilayers, their coupling to the interfacial dislocation network in twisted layers remains unresolved. Here we demonstrate that twisting freestanding SrTiO3 layers undergo interfacial reconstruction into a network of screw dislocations, accompanied by the emergence of in-plane topological vortices. Unlike in previous reports, these vortices are associated with the periodicity of the dislocation network rather than with geometric moire patterns. Four-dimensional scanning transmission electron microscopy (4D-STEM) reveals long-range ordered vortex-antivortex arrays with nearly continuous polarisation rotation. A machine-learning interatomic potential, trained on first-principles calculations, together with phase-field modelling, confirms that competing strains within the dislocation network stabilize polar vortex-antivortex pairs and drive the emergence of an electronic superlattice with a well-defined periodicity. Our results establish twist-controlled dislocation networks as a new and versatile route to designing local polar and electronic structures in oxide materials.

cond-mat.mtrl-sci↗

arXiv2026

Importance of Electronic Entropy for Machine Learning Interatomic Potentials

Machine learning interatomic potentials (MLIPs) enable large-scale atomistic simulations but remain challenged in describing mixed-valence materials where charge ordering strongly influences thermodynamic stability. Here we investigate the role of electronic entropy in MLIP structural optimization of the battery cathode material \ce{NaFePO4}. We show that conventional MLIPs fail to reproduce the correct stability of intermediate \ce{Na} concentrations because structural optimization leads to incorrect \ce{Fe^{2+}}/\ce{Fe^{3+}} charge assignments, resulting in erroneous energy ordering and convex-hull predictions. Analysis of magnetic moments during structural optimization reveals that MLIPs are unable to capture electronic entropy associated with charge ordering. To address this limitation, we introduce an approach that embeds charge-state information directly into the MLIP representation by distinguishing between \ce{Fe^{2+}} and \ce{Fe^{3+}} environments during training. Retraining CHGNet, cPaiNN, and MACE with this representation enables accurate structural optimization, correct identification of charge ordering, and improved agreement with density functional theory convex hulls. Our results demonstrate that incorporating electronic entropy into MLIP representations is essential for modeling charge-disordered materials and provide a practical framework for extending MLIP simulations to mixed-valence transition-metal systems.

cond-mat.mtrl-sci↗

arXiv2026

Reinforcement Learning for Chemical Ordering in Alloy Nanoparticles

We approach the search for optimal element ordering in bimetallic alloy nanoparticles (NPs) as a reinforcement learning (RL) problem and have built an RL agent that learns to perform such global optimization using the geometric graph representation of the NPs. To demonstrate the effectiveness, we train an RL agent to perform composition-conserving atomic swap actions on the icosahedral nanoparticle structure. Trained once on randomized $Ag_{X}Au_{309-X}$ compositions and orderings, the agent discovers previously established ground state structure. We show that this optimization is robust to differently ordered initialisations of the same NP compositions. We also demonstrate that a trained policy can extrapolate effectively to NPs of unseen size. However, the efficacy is limited when multiple alloying elements are involved. Our results demonstrate that RL with pre-trained equivariant graph encodings can navigate combinatorial ordering spaces at the nanoparticle scale, and offer a transferable optimization strategy with the potential to generalize across composition and reduce repeated individual search cost.

cond-mat.mtrl-sci↗

arXiv2026

ELECTRA: A Cartesian Network for 3D Charge Density Prediction with Floating Orbitals

We present the Electronic Tensor Reconstruction Algorithm (ELECTRA) - an equivariant model for predicting electronic charge densities using floating orbitals. Floating orbitals are a long-standing concept in the quantum chemistry community that promises more compact and accurate representations by placing orbitals freely in space, as opposed to centering all orbitals at the position of atoms. Finding the ideal placement of these orbitals requires extensive domain knowledge, though, which thus far has prevented widespread adoption. We solve this in a data-driven manner by training a Cartesian tensor network to predict the orbital positions along with orbital coefficients. This is made possible through a symmetry-breaking mechanism that is used to learn position displacements with lower symmetry than the input molecule while preserving the rotation equivariance of the charge density itself. Inspired by recent successes of Gaussian Splatting in representing densities in space, we are using Gaussian orbitals and predicting their weights and covariance matrices. Our method achieves a state-of-the-art balance between computational efficiency and predictive accuracy on established benchmarks. Furthermore, ELECTRA is able to lower the compute time required to arrive at converged DFT solutions - initializing calculations using our predicted densities yields an average 50.72 % reduction in self-consistent field (SCF) iterations on unseen molecules.

cs.LG↗

arXiv2025

GO-Diff: Data-free and amortized global structure optimization

We introduce GO-Diff, a diffusion-based method for global structure optimization that learns to directly sample low-energy atomic configurations without requiring prior data or explicit relaxation. GO-Diff is trained from scratch using a Boltzmann-weighted score-matching loss, leveraging only the known energy function to guide generation toward thermodynamically favorable regions. The method operates in a two-stage loop of self-sampling and model refinement, progressively improving its ability to target low-energy structures. Compared to traditional optimization pipelines, GO-Diff achieves competitive results with significantly fewer energy evaluations. Moreover, by reusing pretrained models across related systems, GO-Diff supports amortized optimization - enabling faster convergence on new tasks without retraining from scratch.

physics.comp-ph↗

arXiv2025

A foundation model for atomistic materials chemistry

Atomistic simulations of matter, especially those that leverage first-principles (ab initio) electronic structure theory, provide a microscopic view of the world, underpinning much of our understanding of chemistry and materials science. Over the last decade or so, machine-learned force fields have transformed atomistic modeling by enabling simulations of ab initio quality over unprecedented time and length scales. However, early ML force fields have largely been limited by: (i) the substantial computational and human effort of developing and validating potentials for each particular system of interest; and (ii) a general lack of transferability from one chemical system to the next. Here we show that it is possible to create a general-purpose atomistic ML model, trained on a public dataset of moderate size, that is capable of running stable molecular dynamics for a wide range of molecules and materials. We demonstrate the power of the MACE-MP-0 model - and its qualitative and at times quantitative accuracy - on a diverse set of problems in the physical sciences, including properties of solids, liquids, gases, chemical reactions, interfaces and even the dynamics of a small protein. The model can be applied out of the box as a starting or "foundation" model for any atomistic system of interest and, when desired, can be fine-tuned on just a handful of application-specific data points to reach ab initio accuracy. Establishing that a stable force-field model can cover almost all materials changes atomistic modeling in a fundamental way: experienced users get reliable results much faster, and beginners face a lower barrier to entry. Foundation models thus represent a step towards democratising the revolution in atomic-scale modeling that has been brought about by ML force fields.

physics.chem-ph↗

arXiv2025

Dis-GEN: Disordered crystal structure generation

A wide range of synthesized crystalline inorganic materials exhibit compositional disorder, where multiple atomic species partially occupy the same crystallographic site. As a result, the physical and chemical properties of such materials are dependent on how the atomic species are distributed among the corresponding symmetrical sites, making them exceptionally challenging to model using computational methods. For this reason, existing generative models cannot handle the complexities of disordered inorganic crystals. To address this gap, we introduce Dis-GEN, a generative model based on an empirical equivariant representation, derived from theoretical crystallography methodology. Dis-GEN is capable of generating symmetry-consistent structures that accommodate both compositional disorder and vacancies. The model is uniquely trained on experimental structures from the Inorganic Crystal Structure Database (ICSD) - the world's largest database of identified inorganic crystal structures. We demonstrate that Dis-GEN can effectively generate disordered inorganic materials while preserving crystallographic symmetry throughout the generation process. This approach provides a critical check point for the systematic exploration and discovery of disordered functional materials, expanding the scope of generative modeling in materials science.

cond-mat.mtrl-sci↗

arXiv2025

Kinetic Langevin Diffusion for Crystalline Materials Generation

Generative modeling of crystalline materials using diffusion models presents a series of challenges: the data distribution is characterized by inherent symmetries and involves multiple modalities, with some defined on specific manifolds. Notably, the treatment of fractional coordinates representing atomic positions in the unit cell requires careful consideration, as they lie on a hypertorus. In this work, we introduce Kinetic Langevin Diffusion for Materials (KLDM), a novel diffusion model for crystalline materials generation, where the key innovation resides in the modeling of the coordinates. Instead of resorting to Riemannian diffusion on the hypertorus directly, we generalize Trivialized Diffusion Model (TDM) to account for the symmetries inherent to crystals. By coupling coordinates with auxiliary Euclidean variables representing velocities, the diffusion process is now offset to a flat space. This allows us to effectively perform diffusion on the hypertorus while providing a training objective that accounts for the periodic translation symmetry of the true data distribution. We evaluate KLDM on both Crystal Structure Prediction (CSP) and De-novo Generation (DNG) tasks, demonstrating its competitive performance with current state-of-the-art models.

cs.LG↗

arXiv2025

Acoustic Classification of Maritime Vessels using Learnable Filterbanks

Reliably monitoring and recognizing maritime vessels based on acoustic signatures is complicated by the variability of different recording scenarios. A robust classification framework must be able to generalize across diverse acoustic environments and variable source-sensor distances. To this end, we present a deep learning model with robust performance across different recording scenarios. Using a trainable spectral front-end and temporal feature encoder to learn a Gabor filterbank, the model can dynamically emphasize different frequency components. Trained on the VTUAD hydrophone recordings from the Strait of Georgia, our model, CATFISH, achieves a state-of-the-art 96.63 % percent test accuracy across varying source-sensor distances, surpassing the previous benchmark by over 12 percentage points. We present the model, justify our architectural choices, analyze the learned Gabor filters, and perform ablation studies on sensor data fusion and attention-based pooling.

cs.SD↗

arXiv2023

Reducing Training Data Needs with Minimal Multilevel Machine Learning (M3L)

For many machine learning applications in science, data acquisition, not training, is the bottleneck even when avoiding experiments and relying on computation and simulation. Correspondingly, and in order to reduce cost and carbon footprint, training data efficiency is key. We introduce minimal multilevel machine learning (M3L) which optimizes training data set sizes using a loss function at multiple levels of reference data in order to minimize a combination of prediction error with overall training data acquisition costs (as measured by computational wall-times). Numerical evidence has been obtained for calculated atomization energies and electron affinities of thousands of organic molecules at various levels of theory including HF, MP2, DLPNO-CCSD(T), DFHFCABS, PNOMP2F12, and PNOCCSD(T)F12, and treating tens with basis sets TZ, cc-pVTZ, and AVTZ-F12. Our M3L benchmarks for reaching chemical accuracy in distinct chemical compound sub-spaces indicate substantial computational cost reductions by factors of $\sim$ 1.01, 1.1, 3.8, 13.8 and 25.8 when compared to heuristic sub-optimal multilevel machine learning (M2L) for the data sets QM7b, QM9$^\mathrm{LCCSD(T)}$, EGP, QM9$^\mathrm{CCSD(T)}_\mathrm{AE}$, and QM9$^\mathrm{CCSD(T)}_\mathrm{EA}$, respectively. Furthermore, we use M2L to investigate the performance for 76 density functionals when used within multilevel learning and building on the following levels drawn from the hierarchy of Jacobs Ladder:~LDA, GGA, mGGA, and hybrid functionals. Within M2L and the molecules considered, mGGAs do not provide any noticeable advantage over GGAs. Among the functionals considered and in combination with LDA, the three on average top performing GGA and Hybrid levels for atomization energies on QM9 using M3L correspond respectively to PW91, KT2, B97D, and $τ$-HCTH, B3LYP$\ast$(VWN5), TPSSH.

physics.chem-ph↗

arXiv2023

Wave function Ansatz (but Periodic) Networks and the Homogeneous Electron Gas

We design a neural network Ansatz for variationally finding the ground-state wave function of the Homogeneous Electron Gas, a fundamental model in the physics of extended systems of interacting fermions. We study the spin-polarised and paramagnetic phases with 7, 14 and 19 electrons over a broad range of densities from $r_s=1$ to $r_s=100$, obtaining similar or higher accuracy compared to a state-of-the-art iterative backflow baseline even in the challenging regime of very strong correlation. Our work extends previous applications of neural network Ansätze to molecular systems with methods for handling periodic boundary conditions, and makes two notable changes to improve performance: splitting the pairwise streams by spin alignment and generating backflow coordinates for the orbitals from the network. We illustrate the advantage of our high quality wave functions in computing the reduced single particle density matrix. This contribution establishes neural network models as flexible and high precision Ansätze for periodic electronic systems, an important step towards applications to crystalline solids.

physics.chem-ph↗

arXiv2022

NeuralNEB -- Neural Networks can find Reaction Paths Fast

Quantum mechanical methods like Density Functional Theory (DFT) are used with great success alongside efficient search algorithms for studying kinetics of reactive systems. However, DFT is prohibitively expensive for large scale exploration. Machine Learning (ML) models have turned out to be excellent emulators of small molecule DFT calculations and could possibly replace DFT in such tasks. For kinetics, success relies primarily on the models capability to accurately predict the Potential Energy Surface (PES) around transition-states and Minimal Energy Paths (MEPs). Previously this has not been possible due to scarcity of relevant data in the literature. In this paper we train state of the art equivariant Graph Neural Network (GNN)-based models on around 10.000 elementary reactions from the Transition1x dataset. We apply the models as potentials for the Nudged Elastic Band (NEB) algorithm and achieve a Mean Average Error (MAE) of 0.13+/-0.03 eV on barrier energies on unseen reactions. We compare the results against equivalent models trained on QM9 and ANI1x. We also compare with and outperform Density Functional based Tight Binding (DFTB) on both accuracy and computational resource. The implication is that ML models, given relevant data, are now at a level where they can be applied for downstream tasks in quantum chemistry transcending prediction of simple molecular features.

physics.comp-ph↗

arXiv2022

Transition1x -- a Dataset for Building Generalizable Reactive Machine Learning Potentials

Machine Learning (ML) models have, in contrast to their usefulness in molecular dynamics studies, had limited success as surrogate potentials for reaction barrier search. It is due to the scarcity of training data in relevant transition state regions of chemical space. Currently, available datasets for training ML models on small molecular systems almost exclusively contain configurations at or near equilibrium. In this work, we present the dataset Transition1x containing 9.6 million Density Functional Theory (DFT) calculations of forces and energies of molecular configurations on and around reaction pathways at the wB97x/6-31G(d) level of theory. The data was generated by running Nudged Elastic Band (NEB) calculations with DFT on 10k reactions while saving intermediate calculations. We train state-of-the-art equivariant graph message-passing neural network models on Transition1x and cross-validate on the popular ANI1x and QM9 datasets. We show that ML models cannot learn features in transition-state regions solely by training on hitherto popular benchmark datasets. Transition1x is a new challenging benchmark that will provide an important step towards developing next-generation ML force fields that also work far away from equilibrium configurations and reactive systems.

physics.chem-ph↗

arXiv2022

Equivariant graph neural networks for fast electron density estimation of molecules, liquids, and solids

Electron density $ρ(\vec{r})$ is the fundamental variable in the calculation of ground state energy with density functional theory (DFT). Beyond total energy, features and changes in $ρ(\vec{r})$ distributions are often used to capture critical physicochemical phenomena in functional materials. We present a machine learning framework for the prediction of $ρ(\vec{r})$. The model is based on equivariant graph neural networks and the electron density is predicted at special query point vertices that are part of the message passing graph, but only receive messages. The model is tested across multiple data sets of molecules (QM9), liquid ethylene carbonate electrolyte (EC) and LixNiyMnzCo(1-y-z)O2 lithium ion battery cathodes (NMC). For QM9 molecules, the accuracy of the proposed model exceeds typical variability in $ρ(\vec{r})$ obtained from DFT done with different exchange-correlation functionals. The accuracy on all three datasets is beyond state of the art and the computation time is orders of magnitude faster than DFT.

physics.comp-ph↗