Scientific Machine Learning is a new frontier
What is Scientific Machine Learning
SciML is an approach where scientific based model constraints are introduced into Machine Learning algorithms, allowing prediction of future performance of complex multiscale, multiphysics systems using sparse, high-multi-fidelity, and heterogeneous data
This full day tutorial is a comprehensive overview of building and deploying neural forward and inverse PDE solvers. We will discuss Physics-Informed Neural Networks (PINNs), which are usually dense networks producing point-wise solutions of PDEs, as well as CNN-based networks for producing full-field solutions of parametric PDEs, along with GAN-based networks that solve both forward and inverse ODEs.
We will work through several cloud scalable approaches including those for simulating across parameters, as well as distributed deep-learning to obtain high-fidelity solutions. This hands-on tutorial will provide the theoretical background, computational training and software tools for practitioners to rapidly deploy cloud scalable solutions for their forward and inverse PDE needs.
We will provide access to Microsoft Azure HPC clusters, and scripts for attendees to follow all demos at their own pace. We envision this tutorial to be of significant utility for academic and industry practitioners interested in the seamless scale-up of their approaches.
Schedule for Tutorial
Distributed Deep Learning Technical Seminar
The seminar covers fundamentals and different steps in deep learning training and their respective memory footprints and time complexities. We will provide an overview of I/O, memory, and network architectures of CPU and GPU HPC clusters, and describe different modes of parallelism on different hardware infrastructures:
a) Data parallelism on CPU and GPU HPC clusters,
b) Model parallelism on CPU and GPU HPC clusters,
c) Pipeline parallelism, and
d) Federated learning on IOT devices.
Recent advancements in computation and sensor technology have enabled the cheap collection of high-resolution phenotype data across a large geographical area with high temporal resolution. Continuous increase in the amount of data collected and annotated has made it possible to apply deep learning algorithms successfully in a wide variety of challenging plant phenotyping tasks like in-field plant segmentation. The size of an individual image and the number of such images necessitates using large deep learning models. Naturally, distributed deep learning becomes an essential tool for training and deploying such models. We show examples of using distributed training for instance-segmentation on large-scale in-field Maize data, and anomaly detection using a low-dimensional representation of infield Maize data using convolutional autoencoders
Seismic Facies Classification
Identification of different geological features from seismic data by expert seismic interpreters for exploration is often referred to as seismic facies classification. Most of the work related to this problem utilizes seismic facies classification using 2D seismic cross-sections. Seismic facies classifications from 2D cross-sections, when stitched together and analyzed in a 3D holistic view, show abrupt discontinuity of geological features that are unrealistic. Depending on the direction in which the 2D cross-sections are taken, some features might not be fully visible in those sections which leads to wrong interpretations. Here, we use 3D image segmentation models to solve the problem of 3D seismic facies classification. This introduces two challenges: 1) memory requirements in our computational framework and 2) neural network design due to the large compute time to train each of these models. We will present distributed deep learning applications for 3D seismic facies classification problems.
Over the past few decades, there has been much emphasis on designing components with optimal performance to adapt to increasingly competitive markets. Further, coupled with advances in additive manufacturing and other advanced manufacturing processes, the scope to improve components’ performance using design optimization has increased drastically. However, state-of-the-art design optimization frameworks are compute-intensive due to the requirement of performing several iterations of finite element analysis. This part of the tutorial will explore a deep learning-based framework for performing faster and less computational design topology optimization. We will draw parallels between this problem and image segmentation task and then present the application of distributed deep learning for 3D topology optimization at different 3D voxel resolutions using different model architectures and different optimizers including higher-order methods.
Driver behavior has been an important subject studied to improve driver safety and develop intelligent driver-assist systems. Naturalistic driving studies (NDS) are the most sought-after method
that provides insight into the driver’s everyday behavior. In these studies, vehicles are fitted with multiple sensors to record the driver’s actions such as speed, acceleration, and braking and cameras to record events inside and outside the vehicle in real-time. For a human it is easy to watch these videos and accurately identify the cues such as the driver was distracted because of a call which led to a lane departure during heavy traffic conditions. The challenge is to automate the processing using Deep Models. We will present how distributed training is used for simultaneously processing short clip of videos from multiple cameras for traffic conflict classification.
Data Scientist, ML engineer
Post Doc, ISU
PhD Student, ISU
Data Scientist, ML engineer