Introduction to Deep Learning for Scientific Applications
Introduction to Deep Learning for Scientific Applications
Virtual Event
October 7-8, 2020
11:00 AM – 5:15 PM (ET)
REGISTRATION DEADLINE: September 27, 2020
On October 7-8, 2020, the Oak Ridge Leadership Computing Facility (OLCF), in conjunction with IBM, will host a 2-day (virtual) training titled “Introduction to Deep Learning for Scientific Applications”. The goal of the training is to show current Summit users how they might use machine learning (ML) / deep learning (DL) in their traditional (modeling and simulation) scientific applications.
During the event
During the event, IBM and OLCF staff will give an overview of ML/DL concepts and describe how these concepts are relevant to traditional HPC applications. They will start with the basics of ML on the first day and work toward integrating DL surrogates into an application by the end of the second day. The material covered in the presentations will be supplemented by hands-on labs, where participants will have an opportunity to test their understanding of the material. For more details on the topics that will be covered, please see the “Agenda” dropdown below.
If you plan to attend, please make sure to register below.
If you have any questions, please contact Tom Papatheodore ([email protected]).
[tw-toggle title=”Registration”] [/tw-toggle] [tw-toggle title=”Agenda”]
DRAFT AGENDA – SUBJECT TO CHANGE
Day 1 – October 7, 2020
| Time (all times EDT) | Topic | Presenter |
|---|---|---|
| 11:00 AM – 11:15 AM | Welcome and Introduction | Suzanne Parete-Koon OLCF HPC Engineer |
| 11:15 AM – 11:45 AM | A Brief Overview of ML/DL | Clarisse Taaffe-Hedglin- IBM Executive Client Technical Architect |
| 11:45 AM – 12:15 PM | Intersection of AI and HPC | Clarisse Taaffe-Hedglin- IBM Executive Client Technical Architect |
| 12:15 PM – 12:30 PM | Lab 0: Environment Setup for Following Labs | Dustin VanStee- IBM Data Scientist/ AI Solution Architect |
| 12:30 PM – 12:45 PM | Summit System Overview | Tom Paptheodore- OLCF HPC Engineer |
| 12:45 PM – 1:15 PM | Python Environments on Summit | Matt Belllhorn- OLCF HPC Engineer |
| 1:15 PM – 1:30 PM | Break | |
| 1:30 PM – 2:00 PM | Introduction to IBM WML-CE |
Brad Nemanich – IBM Cognitive Software Developer
|
| 2:00 PM – 3:30 PM | Lab 1: Python AI Essentials for Beginners | Travis Siegfried- IBM AI & Vision Solution Architect |
| 3:30 PM – 3:45 PM | Break | |
| 3:45 PM – 5:15 PM | Lab 2: GPU accelerated dataframes – RAPIDS / Dask Lab | Dustin VanStee- IBM Data Scientist/ AI Solution Architect |
Day 2 – October 8, 2020
| Time (all times EDT) | Topic | Presenter |
|---|---|---|
| 11:00 AM – 11:15 AM | Welcome Back | Suzanne Parete-Koon- OLCF HPC Engineer |
| 11:15 AM – 12:00 PM | Overview of Neural Networks | Clarisse Taaffe-Hedglin – IBM Executive Client Technical Architect |
| 12:00 PM – 1:45 PM | Lab 3: Intro to PyTorch (w/lecture) | Dustin VanStee- IBM Data Scientist/ AI Solution Architect |
| 1:45 PM – 2:00 PM | Break | |
| 2:00 PM – 2:30 PM | Integrating Trained Models into C++/FORTRAN Applications | Bob Chesebrough – IBM Data Scientist/ AI Solution Architect |
| 2:30 PM – 3:45 PM | Lab 4: End-to-End Deep Learning (Training) – Using Deep Learning as Universal Function Approximator | Bob Chesebrough – IBM Data Scientist/ AI Solution Architect |
| 3:45 PM – 4:00 PM | Break | |
| 4:00 PM – 5:00 PM | Lab 5: End-to-End Deep Learning (Inference) – Function Representation Integrated into Example Application | Bob Chesebrough – IBM Data Scientist/ AI Solution Architect |
| 5:00 PM – 5:15 PM | Closing Remarks | IBM, OLCF |
