arXiv subjects

Rishi Rao

Publications and source records attributed to Rishi Rao.

5 recordsLinked to original sources

arXiv2026

Learning and retrieval for warm-starting charge-self-consistent DFT+DMFT

Charge-self-consistent (CSC) DFT+DMFT delivers quantitative correlated-electron physics one configuration at a time, making ensemble sampling dependent on reliable warm starts for its expensive fixed-point iteration. Here we compare retrieval from the most similar converged structure with an E(3)-equivariant model that predicts a physics-structured self-energy and Fermi level. Because the full CSC loop refines both initializations, every production result remains a converged DFT+DMFT solution. Across metallic Fe, correlated FeO, and Mott-insulating NiO, learning reduces the typical iterations to sustained convergence from 8 to 3, 8 to 3, and 6 to 1. Retrieval matches this median speed when a dense same-state archive is available, but several poor transplants reveal that structural similarity does not predict transplant quality. In a pre-registered volume window excluded from training and donor pools, learning retains its speed, whereas retrieval starts $3.4\times$ farther from the fixed point and approaches cold-start cost. Either initialization can select a distinct near-degenerate branch on a rugged CSC landscape. Applied end to end, the workflow generates over one thousand correlated energy and force labels for iron at Earth's-core conditions and trains an equivariant interatomic potential. Solid--liquid coexistence with 9216 atoms gives $T_m = 6225 \pm 42\,\mathrm{K}_{\mathrm{stat}}$ at 330~GPa, consistent with recent experiments. A 50-configuration DFT+DMFT audit resolves the potential's energy calibration but does not justify a corrected melting temperature, because four-atom cells cannot realize a liquid. The resulting regime map favors retrieval within dense coverage, amortized learning at and beyond its boundary, and solver refinement throughout.

cond-mat.mtrl-sci↗

arXiv2026

Dataset-aware entropy-maximized active learning for machine-learned interatomic potentials

We present an active learning framework for efficiently generating training data for machine-learned interatomic potentials (MLIPs). The method combines local entropy-driven molecular dynamics with global dataset-aware filtering: a per-configuration entropy term biases MD trajectories toward structurally diverse snapshots, while a global entropy measure, the log-determinant of the fingerprint covariance matrix of the entire dataset, selects only those configurations that provide genuinely new information. We employ dual covariance modes (per-atom for disordered structures and per-config for ordered phases) to achieve broad coverage of configuration space. Combined with a pre-trained foundation model (Allegro-OAM-L) and analytical fingerprint gradients from Gaussian overlap matrix eigenvalues, the framework produces high-quality domain-specific potentials with near- or sub-meV/atom accuracy on test data drawn from the same distribution at training-set sizes of order $10^{2}$ to $10^{3}$ entropy-selected DFT-labeled structures. We demonstrate the method on three systems spanning diverse bonding types and pressure-driven phase transitions: carbon (covalent), silicon (covalent/metallic), and NaCl (ionic). In learning curve comparisons against random molecular dynamics sampling at matched training set sizes ($N = 100$ to $800$), evaluated over three independent training-set draws per condition, entropy-driven sampling achieves a factor of approximately $3$ to $10$ lower energy MAE at $N = 800$ on in-distribution holdouts across the three systems, with the magnitude of the gain depending on the bonding type and the size at which the random-MD baseline saturates.

cond-mat.mtrl-sci↗

arXiv2024

Phase transitions of correlated systems from graph neural networks with quantum embedding techniques

Correlated systems represent a class of materials that are difficult to describe through traditional electronic structure methods. The computational demand to simulate the structural dynamics of such systems, with correlation effects considered, is substantial. Here, we investigate the structural dynamics of $f$- and $d$-electron correlated systems by integrating quantum embedding techniques with interatomic potentials derived from graph neural networks. For Cerium, a prototypical correlated $f$-electron system, we use Density Functional Theory with the Gutzwiller approximation to generate training data due to efficiency with which correlations effects are included for large multi-orbital systems. For Nickel Oxide, a prototypical correlated $d$-electron system, advancements in computational capabilities now permit the use of full Dynamical Mean Field Theory to obtain energies and forces. We train neural networks on this data to create a model of the potential energy surface, enabling rapid and effective exploration of structural dynamics. Utilizing these potentials, we delineate transition pathways between the $α$, $α'$, and $α''$ phases of Cerium and predict the melting curve of Nickel Oxide. Our results demonstrate the potential of machine learning potentials to accelerate the study of strongly correlated systems, offering a scalable approach to explore and understand the complex physics governing these materials.

cond-mat.str-el↗

arXiv2023

Predicting New Heavy Fermion Materials within Carbon-Boron Clathrate Structures

Heavy fermion materials have been a rich playground for strongly correlated physics for decades. However, engineering tunable and synthesizable heavy fermion materials remains a challenge. We strive to integrate heavy fermion properties into carbon boron clathrates as a universal structure which can host a diverse array of interesting physical phenomena. Using a combination of density functional theory and dynamical mean field theory, we study two rare earth carbon boron clathrates, SmB$_3$C$_3$ and CeB$_3$C$_3$, and explore properties arising from the strong electronic correlations. We find a significant increase in the density of states at the Fermi level in CeB$_3$C$_3$ as the temperature is lowered, indicating the development of a heavy electron state. In SmB$_3$C$_3$, a potential Kondo insulating state is identified. Both findings point to rare earth carbon boron clathrates as novel strongly correlated materials within a universally tunable structure, offering a fresh platform to innovate upon conventional heavy-fermion materials design.

cond-mat.mtrl-sci↗

arXiv2022

Can you even tell left from right? Presenting a new challenge for VQA

Visual Question Answering (VQA) needs a means of evaluating the strengths and weaknesses of models. One aspect of such an evaluation is the evaluation of compositional generalisation, or the ability of a model to answer well on scenes whose scene-setups are different from the training set. Therefore, for this purpose, we need datasets whose train and test sets differ significantly in composition. In this work, we present several quantitative measures of compositional separation and find that popular datasets for VQA are not good evaluators. To solve this, we present Uncommon Objects in Unseen Configurations (UOUC), a synthetic dataset for VQA. UOUC is at once fairly complex while also being well-separated, compositionally. The object-class of UOUC consists of 380 clasess taken from 528 characters from the Dungeons and Dragons game. The train set of UOUC consists of 200,000 scenes; whereas the test set consists of 30,000 scenes. In order to study compositional generalisation, simple reasoning and memorisation, each scene of UOUC is annotated with up to 10 novel questions. These deal with spatial relationships, hypothetical changes to scenes, counting, comparison, memorisation and memory-based reasoning. In total, UOUC presents over 2 million questions. UOUC also finds itself as a strong challenge to well-performing models for VQA. Our evaluation of recent models for VQA shows poor compositional generalisation, and comparatively lower ability towards simple reasoning. These results suggest that UOUC could lead to advances in research by being a strong benchmark for VQA.

cs.CV↗