no code implementations • 24 Apr 2024 • Akshay Thakur, Souvik Chakraborty
We propose a neural operator framework, termed mixture density nonlinear manifold decoder (MD-NOMAD), for stochastic simulators.
no code implementations • 24 Apr 2024 • Sawan Kumar, Rajdip Nayek, Souvik Chakraborty
The study of neural operators has paved the way for the development of efficient approaches for solving partial differential equations (PDEs) compared with traditional methods.
1 code implementation • 24 Feb 2024 • Harshil Vagadia, Mudit Chopra, Abhinav Barnawal, Tamajit Banerjee, Shreshth Tuli, Souvik Chakraborty, Rohan Paul
PhyPlan leverages PINNs to simulate and predict outcomes of actions in a fast and accurate manner and uses MCTS for planning.
no code implementations • 8 Jan 2024 • Jyoti Rani, Tapas Tripura, Hariprasad Kodamana, Souvik Chakraborty
This article proposes a generative adversarial wavelet neural operator (GAWNO) as a novel unsupervised deep learning approach for fault detection and isolation of multivariate time series processes. The GAWNO combines the strengths of wavelet neural operators and generative adversarial networks (GANs) to effectively capture both the temporal distributions and the spatial dependencies among different variables of an underlying system.
no code implementations • 15 Nov 2023 • Shailesh Garg, Souvik Chakraborty
We propose, in this paper, a Variable Spiking Wavelet Neural Operator (VS-WNO), which aims to bridge the gap between theoretical and practical implementation of Artificial Intelligence (AI) algorithms for mechanics applications.
no code implementations • 15 Nov 2023 • Shailesh Garg, Souvik Chakraborty
This property of the proposed VSN makes it suitable for regression tasks, which is a weak point for the vanilla spiking neurons, all while keeping the energy budget low.
no code implementations • 29 Oct 2023 • Tapas Tripura, Souvik Chakraborty
The proposed foundational model offers two key advantages: (i) it can simultaneously learn solution operators for multiple parametric PDEs, and (ii) it can swiftly generalize to new parametric PDEs with minimal fine-tuning.
no code implementations • 10 Oct 2023 • Tapas Tripura, Souvik Chakraborty
Unlike existing neural network-based approaches, the proposed approach (a) yields an interpretable description of Lagrangian, (b) exploits Bayesian learning to quantify the epistemic uncertainty due to limited data, (c) automates the distillation of Hamiltonian from the learned Lagrangian using Legendre transformation, and (d) provides ordinary (ODE) and partial differential equation (PDE) based descriptions of the observed systems.
no code implementations • 8 Oct 2023 • N Navaneeth, Souvik Chakraborty
In this work, we propose "waveformer", a novel operator learning approach for learning solutions of dynamical systems.
no code implementations • 24 Sep 2023 • Tushar, Souvik Chakraborty
The well-known governing physics in science and engineering is often based on certain assumptions and approximations.
no code implementations • 28 Jun 2023 • Yogesh Chandrakant Mathpati, Tapas Tripura, Rajdip Nayek, Souvik Chakraborty
We propose a novel framework for discovering Stochastic Partial Differential Equations (SPDEs) from data.
no code implementations • 8 Jun 2023 • Kalpesh More, Tapas Tripura, Rajdip Nayek, Souvik Chakraborty
To accelerate the overall process, a variational Bayes-based approach for discovering partial differential equations is proposed.
no code implementations • 12 Feb 2023 • Navaneeth N, Tapas Tripura, Souvik Chakraborty
Deep neural operators are recognized as an effective tool for learning solution operators of complex partial differential equations (PDEs).
no code implementations • 9 Feb 2023 • Tapas Tripura, Souvik Chakraborty
The Lagrangian are derived in interpretable forms, which also allows the automated discovery of conservation laws and governing equations of motion.
no code implementations • 2 Feb 2023 • Shailesh Garg, Souvik Chakraborty
In this paper, we propose a novel data-driven operator learning framework referred to as the \textit{Randomized Prior Wavelet Neural Operator} (RP-WNO).
no code implementations • 17 Jan 2023 • Kazuma Kobayashi, Bader Almutairi, Md Nazmus Sakib, Souvik Chakraborty, Syed B. Alam
Therefore, the use of Explainable AI (XAI) and interpretable machine learning (IML) is crucial for the accurate prediction of prognostics, such as remaining useful life (RUL) in a digital twin system to make it intelligent while ensuring that the AI model is transparent in its decision-making processes and that the predictions it generates can be understood and trusted by users.
Decision Making Explainable Artificial Intelligence (XAI) +1
no code implementations • 19 Dec 2022 • Tapas Tripura, Aarya Sheetal Desai, Sondipon Adhikari, Souvik Chakraborty
A framework for creating and updating digital twins for dynamical systems from a library of physics-based functions is proposed.
no code implementations • 13 Dec 2022 • Yogesh Chandrakant Mathpati, Kalpesh Sanjay More, Tapas Tripura, Rajdip Nayek, Souvik Chakraborty
A two-stage approach is adopted: in the first stage, an efficient variational Bayesian equation discovery algorithm is developed to determine the governing physics of an underlying stochastic differential equation (SDE) from measured output data.
no code implementations • 23 Nov 2022 • James Daniell, Kazuma Kobayashi, Susmita Naskar, Dinesh Kumar, Souvik Chakraborty, Ayodeji Alajo, Ethan Taber, Joseph Graham, Syed Alam
In order to address this gap, this study specifically focuses on the "ML-driven prediction algorithms" as a viable component for the nuclear reactor operation while assessing the reliability and efficacy of the proposed model.
no code implementations • 23 Nov 2022 • Tapas Tripura, Souvik Chakraborty
The proposed approach first discovers \textit{interpretable} governing differential equations from data using a novel algorithm and blends it with a model predictive control algorithm.
no code implementations • 14 Oct 2022 • Kazuma Kobayashi, Dinesh Kumar, Matthew Bonney, Souvik Chakraborty, Kyle Paaren, Syed Alam
To understand the potential of intelligent confirmatory tools, the U. S. Nuclear Regulatory Committee (NRC) initiated a future-focused research project to assess the regulatory viability of machine learning (ML) and artificial intelligence (AI)-driven Digital Twins (DTs) for nuclear power applications.
no code implementations • 20 Sep 2022 • Tushar, Souvik Chakraborty
The primary idea here is to exploit DNN to model the missing physics.
no code implementations • 11 Aug 2022 • Tapas Tripura, Souvik Chakraborty
The existing techniques for equations discovery are dependent on both input and state measurements; however, in practice, we only have access to the output measurements only.
no code implementations • 11 Aug 2022 • Akshay Thakur, Tapas Tripura, Souvik Chakraborty
However, this issue can be alleviated with the use of multi-fidelity learning, where a model is trained by making use of a large amount of inexpensive low-fidelity data along with a small amount of expensive high-fidelity data.
no code implementations • 31 Jul 2022 • Siladittya Manna, Rakesh Dey, Souvik Chakraborty
Supervised Learning algorithms require a large volumes of balanced data to learn robust representations.
no code implementations • 12 Jun 2022 • Shailesh Garg, Souvik Chakraborty
Neural network based data-driven operator learning schemes have shown tremendous potential in computational mechanics.
no code implementations • 4 May 2022 • Tapas Tripura, Souvik Chakraborty
With massive advancements in sensor technologies and Internet-of-things, we now have access to terabytes of historical data; however, there is a lack of clarity in how to best exploit the data to predict future events.
no code implementations • 12 Mar 2022 • Yash Kumar, Souvik Chakraborty
Based on this technique we present two models for discrete and continuous prediction in space.
no code implementations • 5 Mar 2022 • Navaneeth N., Souvik Chakraborty
Unlike purely data-driven approaches, the proposed approach is robust even in the presence of uncertainties; this renders the proposed approach suitable for time-dependent reliability analysis.
no code implementations • 31 Jan 2022 • Shailesh Garg, Harshit Gupta, Souvik Chakraborty
Time dependent reliability analysis and uncertainty quantification of structural system subjected to stochastic forcing function is a challenging endeavour as it necessitates considerable computational time.
no code implementations • 19 Jan 2022 • Akshay Thakur, Souvik Chakraborty
We propose a novel \textit{capsule} based deep encoder-decoder model for surrogate modeling and uncertainty quantification of systems in mechanics from sparse data.
no code implementations • 8 Nov 2021 • Sai Krishna Mendu, Souvik Chakraborty
The proposed GLU-net treats the uncertainty propagation problem as an image to image regression and hence, is extremely data efficient.
no code implementations • 24 Oct 2021 • Akshay Thakur, Souvik Chakraborty
We propose a deep learning-based surrogate model for stochastic simulators.
no code implementations • 1 Sep 2021 • Shailesh Garg, Souvik Chakraborty, Budhaditya Hazra
For improving the predictive capability of the underlying physics, we first use machine learning algorithm to learn a mapping between the estimated state and the input (model-form error) and then introduce it into the governing equation as an additional term.
no code implementations • 24 Aug 2021 • Yash Kumar, Souvik Chakraborty
With recent developments in the field of artificial intelligence and machine learning, the solution of PDEs using neural network has emerged as a domain with huge potential.
no code implementations • 11 May 2021 • Navaneeth N., Souvik Chakraborty
The basic premise is to train the surrogate model on a low-dimensional manifold, discovered using the active subspace algorithm.
no code implementations • 29 Mar 2021 • Shailesh Garg, Ankush Gogoi, Souvik Chakraborty, Budhaditya Hazra
In this paper, we propose a novel digital twin framework for stochastic nonlinear multi-degree of freedom (MDOF) dynamical systems.
no code implementations • 27 Jan 2021 • Yash Kumar, Pranav Bahl, Souvik Chakraborty
We illustrate that utilizing sequential data allows for state recovery from only one or two sensors.
no code implementations • 19 May 2020 • Souvik Chakraborty
This is followed by transfer learning where the low-fidelity model is updated by using the available high-fidelity data.
no code implementations • 12 May 2020 • Souvik Chakraborty, Sondipon Adhikari
Our approach strategically separates into two components -- (a) a physics-based nominal model for data processing and response predictions, and (b) a data-driven machine learning model for the time-evolution of the system parameters.
no code implementations • 4 May 2020 • Souvik Chakraborty
Moreover, the primary bottleneck of solving reliability analysis problems, i. e., running expensive simulations to generate data, is eliminated with this method.
no code implementations • 25 Jan 2020 • Souvik Chakraborty, Sondipon Adhikari, Ranjan Ganguli
As digital twins are also expected to exploit data and computational methods, there is a compelling case for the use of surrogate models in this context.
no code implementations • 4 Jul 2019 • Somdatta Goswami, Cosmin Anitescu, Souvik Chakraborty, Timon Rabczuk
While most of the PINN algorithms available in the literature minimize the residual of the governing partial differential equation, the proposed approach takes a different path by minimizing the variational energy of the system.
no code implementations • 29 Mar 2019 • Rajdip Nayek, Souvik Chakraborty, Sriram Narasimhan
A novel methodology using Gaussian process latent force models is proposed to tackle the problem in a stochastic setting.