Search Results for author: Jay Pathak

Found 7 papers, 1 papers with code

Geometry encoding for numerical simulations

1 code implementation15 Apr 2021 Amir Maleki, Jan Heyse, Rishikesh Ranade, Haiyang He, Priya Kasimbeg, Jay Pathak

We present a notion of geometry encoding suitable for machine learning-based numerical simulation.

One-shot learning for solution operators of partial differential equations

no code implementations6 Apr 2021 Lu Lu, Haiyang He, Priya Kasimbeg, Rishikesh Ranade, Jay Pathak

Discovering governing equations of a physical system, represented by partial differential equations (PDEs), from data is a central challenge in a variety of areas of science and engineering.

One-Shot Learning

A Latent space solver for PDE generalization

no code implementations6 Apr 2021 Rishikesh Ranade, Chris Hill, Haiyang He, Amir Maleki, Jay Pathak

In this work we propose a hybrid solver to solve partial differential equation (PDE)s in the latent space.

ActivationNet: Representation learning to predict contact quality of interacting 3-D surfaces in engineering designs

no code implementations21 Mar 2021 Rishikesh Ranade, Jay Pathak

In machine learning applications, 3-D surfaces are most suitably represented with point clouds or meshes and learning representations of interacting geometries form point-based representations is challenging.

Representation Learning

Physics-consistent deep learning for structural topology optimization

no code implementations9 Dec 2020 Jaydeep Rade, Aditya Balu, Ethan Herron, Jay Pathak, Rishikesh Ranade, Soumik Sarkar, Adarsh Krishnamurthy

However, current state-of-the-art topology optimization frameworks are compute-intensive, mainly due to multiple finite element analysis iterations required to evaluate the component's performance during the optimization process.

An unsupervised learning approach to solving heat equations on chip based on Auto Encoder and Image Gradient

no code implementations19 Jul 2020 Haiyang He, Jay Pathak

Specifically, a hybrid framework of Auto Encoder (AE) and Image Gradient (IG) based network is designed.

DiscretizationNet: A Machine-Learning based solver for Navier-Stokes Equations using Finite Volume Discretization

no code implementations17 May 2020 Rishikesh Ranade, Chris Hill, Jay Pathak

The two solver characteristics that have been adopted in this work are: 1) the use of discretization-based schemes to approximate spatio-temporal partial derivatives and 2) the use of iterative algorithms to solve linearized PDEs in their discrete form.

Cannot find the paper you are looking for? You can Submit a new open access paper.