NVIDIA Deep Learning Platform and latest information by Takeshi Izaki, Nvidia - TMLS #4

Deep Learning has been evolved rapidly recent years. In particular it has started being utilized in various fields over the past year. Today’s Deep Learning solution relies almost exclusively on NVIDIA Deep Learning Platform including GPU. In this session I will talk about overview of this platform and some latest information.

Transcript

1. Senior Manager Takeshi Izaki NVIDIA, Deep Learning Group NVIDIA DEEP LEARNING PLATFORM AND LATEST INFORMATION
2. An amazing year in ai AlphaGo Rivals a World Champion Microsoft & Google “Superhuman” Image Recognition Microsoft “Super Deep Network” Berkeley’s Brett One network, everything robotics Deep Speech 2 One network, 2 languages A New Computing Model Hits Pop Culture
3. A new computing model 0% 10% 20% 30% 40% 50% 60% 70% 80% 90% 100% 2009 2010 2011 2012 2013 2014 2015 2016 Traditional CV Deep Learning ImageNet Deep Learning Object Detection DNN + Data + HPC Traditional Computer Vision Experts + Time Deep Learning Achieves “Superhuman” Results
4. exabytes of content produced daily 10M Users 40 years of video/day 1.7M Broadcasters Users watch 1.5 hours/day 6B Queries/day 10% use speech 270M Items sold/day 43% on mobile devices 8B Video views/day 400% growth in 6 months 300 hours of video/minute 50% on mobile devices
5. Deep Learning Use Cases Image Classification, Object Detection, Localization Face Recognition Speech & Natural Language Processing Medical Imaging & Interpretation Autonomous Machines Recommendation
6. Why is Deep learning hot now? “Google’s AI engine also reflects how the world of computer hardware is changing. (It) depends on machines equipped with GPUs… And it depends on these chips more than the larger tech universe realizes.” DNN GPU BIG DATA
7. Ex. Typical network Many Matrix Multiply with many training data Purpose Face recognition Training Data 10M～100M Images Network Architecture 10 Layers 1B Parameters Learning Algorithm ~30 ExaFLOPS ~30 GPU Days
8. 8 Machine Learning evolution by NVIDIA TESLA GPU GOOGLE BRAIN APPLICATION – DEEP LEARNING BEFORE TESLA AFTER TESLA Cost $5,000K $200K Server# 1,000 server 16 Tesla server Power 600 KW 4 KW Performace 1x 6x
9. NVIDIA GPU for hyperscale 10X Speed up | 20 images/s/W Cloud Services Powered by AI TESLA M40 + TESLA M4
10. TESLA P100 THE MOST ADVANCED HYPERSCALE DATACENTER GPU EVER BUILT 5.3TF FP64 | 10.6TF FP32 | 21.2TF FP16
11. Tesla P100 Advanced Technology 16nm FinFET Pascal Architecture HBM2 Stacked Memory NVLink System Intercnnect
12. NVIDIA DGX-1 World’s First Deep Learning Super Computer Engineered for Deep Learning 170 TF FP16 8 x Tesla P100 NVLink Hybrid Cube Mesh Accelerates major AI frameworks
13. NVIDIA Deep Learning Platform Computer Vision Speech and Sound Behavior Object Detection Voice Recognition Translation Recommendation Engines Sentiment Analysis cuDNN cuBLAS cuSPARSE NCCL cuFFT Mocha.jl Image Classification Deep Learning SDK Framework Application GPU Platform Cloud GPU Tesla P100 Tesla K80/M40/M4 Jetson TX1 Server DGX-1 GIE DRIVEPX2 Deep Learning Math Library Multi-GPU Communication
14. Manufacturing Ref. MONOist Achieved 90% success rate only after 8hours Learning （The same rate as human experts) DEEP LEARNING DAY2016
15. Medical ~ Diagnosis to Prevention http://www.enlitic.com/
16. Finance Algo trading Over +0.5% -0.5%～+0.5% Under -0.5% 1Hour Predict time 14:00 Target time 15:00 Up Down In Range DEEP LEARNING DAY2016
17. ? ? ? In Store Marketing
18. Recommendations Input spectrograms Output “Tag”
19. Q&A System/NLP
20. Product search
21. GTC JAPAN 2016 ！ October 5th (Wed) 2016 Hilton Tokyo Odaiba
22. JEN－HSUN HUANG WILL COME