The International Conference for High Performance Computing, Networking, Storage and Analysis

Session

· Presenters · Organizations · Search Program

Workshop: Machine Learning in HPC Environments

Session Chair

Steven R. Young

Oak Ridge National Laboratory

Event TypeWorkshop

Tags

TimeMonday, November 13th9am - 5:30pm

Location502-503-504

Presentations

9:00am - 9:10am	Introduction - Machine Learning in HPC Environments Author/Presenters Steven R. Young Lawrence Carin Janis Keuper Robert M. Patton Brian Van Essen
9:10am - 10:00am	Leadership AI - Keynote by Jack Wells - Director of Science - Oak Ridge Leadership Computing Facility Author/Presenter Jack Wells
10:00am - 10:30am	Morning Break
10:30am - 11:00am	TensorQuant - A Simulation Toolbox for Deep Neural Network Quantization Author/Presenters Dominik Marek Loroch Norbert Wehn Franz-Josef Pfreundt Janis Keuper
11:00am - 11:30am	An Efficient Task-Based All-Reduce for Machine Learning Applications Author/Presenters Zhenyu Li James Davis Stephen Jarvis
11:30am - 12:00pm	Accelerating Deep Neural Network Learning for Speech Recognition on a Cluster of GPUs Author/Presenter Guojing Cong
12:00pm - 12:30pm	Optimizing Convolutional Neural Networks for Cloud Detection Author/Presenter Travis Johnston
12:30pm - 2:00pm	Lunch (participants on their own)
2:00pm - 2:40pm	Production Deep Learning and Scale - Keynote by Mike Houston - Senior Distinguished Engineer - Deep Learning - Nvidia Author/Presenter Mike Houston
2:40pm - 3:00pm	Training Distributed Deep Recurrent Neural Networks with Mixed Precision on GPU Clusters Author/Presenters Alexey Svyatkovskiy Julian Kates-Harbeck William Tang
3:00pm - 3:30pm	Afternoon Break
3:30pm - 3:54pm	Toward Scalable Parallel Training of Deep Neural Networks Author/Presenters Sam Ade Jacobs Nikoli Dryden Roger Pearce Brian Van Essen
3:54pm - 4:18pm	BlazingText: Scaling and Accelerating Word2Vec using Multiple GPUs Author/Presenter Saurabh Gupta
4:18pm - 4:42pm	An In-Depth Performance Characterization of CPU- and GPU-Based DNN Training on Modern Architectures Author/Presenters Ammar Ahmad Awan Hari Subramoni Dhabaleswar K. Panda
4:42pm - 5:06pm	Designing a Synchronization-Reducing Clustering Method on Manycores: Some Issues and Improvements Author/Presenters Weijian Zheng Fengguang Song Lan Lin
5:06pm - 5:30pm	Evolving Deep Networks Using HPC Author/Presenter Steven Young