21 citations · 27 across the 9 of their papers we have counts for
9 papers
Bayesian 3D Steerable CNNs: Enabling Equivariance and Uncertainty Quantification Simultaneously
Abhishek Keripale, Ponkrshnan Thiagarajan, Susanta Ghosh
Steerable convolutional neural networks (Steerable-CNNs) guarantee SE(3)-equivariance by parameterizing kernels as linear combinations of steerable basis functions, but their deter…
Feature-preserving Latent-EnKF for Data Assimilation of Flows with Shocks
Hemanth Chandravamsi, Hangchuan Hu, Ponkrshnan Thiagarajan +1
The ensemble Kalman filter (EnKF) is widely adopted for sequential data assimilation, but fails for solutions with discontinuities, such as shocks in compressible flows. For ensemb…
Accelerating Hamiltonian Monte Carlo for Bayesian Inference in Neural Networks and Neural Operators
Ponkrshnan Thiagarajan, Tamer A. Zaki, Michael D. Shields
Hamiltonian Monte Carlo (HMC) is a powerful and accurate method to sample from the posterior distribution in Bayesian inference. However, HMC techniques are computationally demandi…
Surprisingly High Redundancy in Electronic Structure Data Across Materials Explained by Low Intrinsic Dimensionality
Sazzad Hossain, Ponkrshnan Thiagarajan, Shashank Pathrudkar +4
Machine learning (ML) models for electronic structure typically rely on large datasets generated by computationally expensive Kohn-Sham density functional theory calculations, as i…
Electronic structure prediction of medium and high entropy alloys across composition space
Shashank Pathrudkar, Stephanie Taylor, Abhishek Keripale +6
We propose machine learning (ML) models to predict the electron density -- the fundamental unknown of a material's ground state -- across the composition space of concentrated allo…
Bayesian Calibration and Uncertainty Quantification of a Rate-dependent Cohesive Zone Model for Polymer Interfaces
Ponkrshnan Thiagarajan, Trisha Sain, Susanta Ghosh
In the present work, a rate-dependent cohesive zone model for the fracture of polymeric interfaces is presented. Inverse calibration of parameters for such complex models through t…
Electronic Structure Prediction of Multi-million Atom Systems Through Uncertainty Quantification Enabled Transfer Learning
Shashank Pathrudkar, Ponkrshnan Thiagarajan, Shivang Agarwal +2
The ground state electron density -- obtainable using Kohn-Sham Density Functional Theory (KS-DFT) simulations -- contains a wealth of material information, making its prediction v…
Jensen-Shannon Divergence Based Novel Loss Functions for Bayesian Neural Networks
Ponkrshnan Thiagarajan, Susanta Ghosh
Bayesian neural networks (BNNs) are state-of-the-art machine learning methods that can naturally regularize and systematically quantify uncertainties using their stochastic paramet…
Explanation and Use of Uncertainty Quantified by Bayesian Neural Network Classifiers for Breast Histopathology Images
Ponkrshnan Thiagarajan, Pushkar Khairnar, Susanta Ghosh
Despite the promise of Convolutional neural network (CNN) based classification models for histopathological images, it is infeasible to quantify its uncertainties. Moreover, CNNs m…