Deep Neural Networks Assisted Simulation Surrogates for Parameter Space Exploration
Deep Neural Networks Assisted Simulation Surrogates for Parameter Space Exploration


For compute-intensive simulation models with high-dimensional input parameters and output spaces, repeated execution of the expensive simulations can become computationally prohibitive and the analysis of the large output is non-trivial. In this talk, I will discuss our recent work on assisting simulation paramter exploration using deep neural networks that won a best paper and an honorable mention awards in the IEEE Visualization conference. To capture the characteristics of
simulation and provide a rapid visualization of the simulation output without huge computation and storage overhead, we develop DNN models to serve as surrogates for prediction of simulation data and visualization images with novel simulation parameters. To achieve these, at simulation time we will collect reduced simulation data and visualization images. Using them as the training data, in one of our recent works we train a generative adversarial networks (GANs) that can take novel simulation and visualization parameters as input to generate new visualization images, which will serve as a quick preview of the simulation output. In another work, we collaborated with computational biologists to design an interactive visual analysis framework, backed by a neural network-based surrogate model that can assist computational biologists in analyzing and visualizing a complex yeast cell polarization simulation model. Our model also allows uncertainty quantification and sensity analysis of the input parameters and simulation output.


Han-Wei Shen is a Full Professor at The Ohio State University. He is currently the Associate Editor in Chief of IEEE Transactions on Visualization and Computer Graphics. He is also the chair of the steering committee for IEEE SciVis conference. His primary research interests are scientific visualization and computer graphics. Professor Shen is a winner of National Science Foundation's CAREER award and US Department of Energy's Early Career Principal Investigator Award. He has served as an Associate Editor for IEEE Transactions on Visualization and Computer Graphics, a paper chair for IEEE Visualization, IEEE Pacific Visualization, and IEEE Parallel Visualization and Graphics. He is currently on the IEEE Visualization conference executive committee, and IEEE SciVis steering committee. He has published more than 50 papers in IEEE Transactions on Visualization and Computer Graphics and IEEE Visualization conference, the very top journal and conference. He received his BS degree from Department of Computer Science and Information Engineering at National Taiwan University in 1988, the MS degree in computer science from the State University of New York at Stony Brook in 1992, and the PhD degree in computer science from the University of Utah in 1998. From 1996 to 1999, he was a research scientist at NASA Ames Research Center in Mountain View California.

Open Data Science




Open Data Science
One Broadway
Cambridge, MA 02142

Privacy Settings
We use cookies to enhance your experience while using our website. If you are using our Services via a browser you can restrict, block or remove cookies through your web browser settings. We also use content and scripts from third parties that may use tracking technologies. You can selectively provide your consent below to allow such third party embeds. For complete information about the cookies we use, data we collect and how we process them, please check our Privacy Policy
Consent to display content from - Youtube
Consent to display content from - Vimeo
Google Maps
Consent to display content from - Google