Session - SC19

DescriptionThe intent of this workshop is to bring together researchers, practitioners, and scientific communities to discuss methods that utilize extreme scale systems for machine learning. This workshop will focus on the greatest challenges in utilizing HPC for machine learning and methods for exploiting data parallelism, model parallelism, ensembles, and parameter search. We invite researchers and practitioners to participate in this workshop to discuss the challenges in using HPC for machine learning and to share the wide range of applications that would benefit from HPC powered machine learning.

https://ornlcda.github.io/MLHPC2019/

Presentations

9:00am - 9:01am	Machine Learning in HPC Environments Organizers Seung-Hwan Lim Xipeng Shen Janis Keuper Michael Houston
9:01am - 9:10am	Opening Remarks: Machine Learning in HPC Environments Presenters Janis Keuper Seung-Hwan Lim Michael Houston Xipeng Shen
9:10am - 10:00am	Morning Keynote : HPC for ML and ML for HPC - scalability, communication, and programming Presenter Torsten Hoefler
10:00am - 10:30am	Machine Learning in HPC Environments Morning Break
10:30am - 11:00am	Understanding Scalability and Fine-Grain Parallelism of Synchronous Data Parallel Training Author/Presenters Jiali Li Bogdan Nicolae Justin Wozniak George Bosilca
11:00am - 11:30am	Fine-Grained Exploitation of Mixed Precision for Faster CNN Training Author/Presenters J. Travis Johnston Steven R. Young Catherine D. Schuman Junghoon Chae Don D. March Robert M. Patton Thomas E. Potok
11:30am - 12:00pm	Metaoptimization on a Distributed System for Deep Reinforcement Learning Author/Presenters Greg Heinrich Iuri Frosio
12:00pm - 12:30pm	Scheduling Optimization of Parallel Linear Algebra Algorithms Using Supervised Learning Author/Presenters Gabriel Laberge Shahrzad Shirzad Patrick Diehl Hartmut Kaiser Serge Prudhomme Adrian S. Lemoine
12:30pm - 2:00pm	Machine Learning in HPC Environments Lunch Break
2:00pm - 2:30pm	Parallel Data-Local Training for Optimizing Word2Vec Embeddings for Word and Graph Embeddings Author/Presenters Gordon E. Moon Denis Newman-Griffis Jinsung Kim Aravind Sukumaran-Rajam Eric Fosler-Lussier P. Sadayappan
2:30pm - 3:00pm	Scalable Hyperparameter Optimization with Lazy Gaussian Processes Author/Presenters Raju Ram Sabine Müller Franz-Josef Pfreundt Nicolas R. Gauger Janis Keuper
3:00pm - 3:30pm	Machine Learning in HPC Environments Afternoon Break
3:30pm - 4:20pm	Afternoon Keynote - Running large models in minutes: an engineering journey through high performance for AI Presenter Julie Bernauer
4:20pm - 4:50pm	GradVis: Visualization and Second Order Analysis of Optimization Surfaces During the Training of Deep Neural Networks Author/Presenters Avraam Chatzimichailidis Janis Keuper Franz-Josef Pfreundt Nicolas R. Gauger
4:50pm - 5:20pm	DisCo: Physics-Based Unsupervised Discovery of Coherent Structures in Spatiotemporal Systems Author/Presenters Adam Rupe Nalini Kumar Vladislav Epifanov Karthik Kashinath Oleksandr Pavlyk Frank Schlimbach Mostofa Patwary Sergey Maidanov Victor Lee Mr Prabhat James P. Crutchfield
5:20pm - 5:30pm	Concluding Remarks - Machine Learning in HPC Environment