SlideShare a Scribd company logo

The world of Machine Learning, Deep Learning and PowerAI

Download as PPTX, PDF
David Spurway, profile picture
David Spurway

The document discusses the advancements and applications of machine learning and deep learning, highlighting the role of IBM's PowerAI in facilitating these technologies. It covers various use cases across industries, challenges faced, and describes the benefits of using deep learning frameworks and high-performance computing. Additionally, it emphasizes the importance of data preparation, model training, and optimization within the machine learning ecosystem.

Technology
Related topics:
Power Systems
1 of 53
Downloaded 61 times
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
© IBM Corporation, 2016 The World of Machine Learning, Deep Learning and PowerAI – Deeper Dive Power Up Live with SCC – 6th June 2017 Presented by David Spurway IBM Power Systems Product Manager IBM Systems, UK and Ireland
2 © IBM Corporation, 2016 Augmented intelligence, Artificial Intelligence, Cognitive driving innovation Faster
3 © IBM Corporation, 2016 My friends at Uni…
4 © IBM Corporation, 2016 From Big Data to AI client journey
5 © IBM Corporation, 2016 OBSERVATION DECISIONINTERPRETATION EVALUATION 010101010101010111100010011001010111 0000000000010101010100000000000 111101011 11000 000000000000 111111 010101 101010 10101010100 Prescriptive Best Outcomes? Descriptive What Has Happened? Cognitive Learn Dynamically Predictive What Could Happen?
6 © IBM Corporation, 2016 010101010101010111100010011001010111 0000000000010101010100000000000 111101011 11000 000000000000 111111 010101 101010 10101010100 OBSERVATION DECISIONINTERPRETATION EVALUATION Prescriptive Best Outcomes? Descriptive What Has Happened? Cognitive Learn Dynamically Predictive What Could Happen? ACTIONDATA How many frauds during last month ? Per Country ? Which Transactions will be fraudulent ? What is the best action in light of potential fraud ? In Natural Language : « Explain me why this transaction is fraudulent ?
7 © IBM Corporation, 2016 Prescriptive Best Outcomes? Descriptive What Has Happened? Cognitive Learn Dynamically Predictive What Could Happen? - Artificial - Intelligence - Big Data - NLP Robot Knowledge Base Deep Learning Machine Learning010101010101010111100010011001010111 1000101 100010 1 1000101 111010111010 00000000000010101010100000000000 111101011
8 © IBM Corporation, 2016 Prepare the data ALL DATA Input VAR Training Data Test Data Machine Learning Algorithms Predictive Model PREDICTION ? Test Data ACCURACY ? Machine Learning Algorithms use training data to create a predictive model: its accuracy is tested on holdback data Machine Learning TensorFlow Caffee Torch Theano Chainer Spark ML …… Prepare Data Build/Train Model Deploy/Score Monitor/Refine Iterative Development
9 © IBM Corporation, 2016 My recent buyer’s journey…
10 © IBM Corporation, 2016 Some challenges… https://uk.pinterest.com/pin/197595502370835426/sent/?sender=5279 06524979412201&invite_code=7fc292525ac44aafa6736bdec95dd1b5 Speziale Floral Lace Fit & Flare Dress Items in this section are temporarily out of stock
11 © IBM Corporation, 2016 Where I ended up going… Petite Clothing Update your wardrobe with Wallis' stunning must have petite range. Designed for women who are 5'3" and under
12 © IBM Corporation, 2016 Deep Learning Goes to the Dogs • https://openpowerfoundation.org/blogs/deep-learning-goes-to-the-dogs/ • http://vision.stanford.edu/aditya86/ImageNetDogs/ • The Stanford Dogs dataset contains images of 120 breeds of dogs from around the world. This dataset has been built using images and annotation from ImageNet for the task of fine-grained image categorization. • https://www.youtube.com/watch?v=6ZRuTWpIo4M
13 © IBM Corporation, 2016 Example of Datasets available http://mmlab.ie.cuhk.edu.hk/projects/DeepFashion.html
14 © IBM Corporation, 2016 DeepFashion: In-shop Clothes Retrieval Details In-shop Clothes Retrieval Benchmark evaluates the performance of in- shop Clothes Retrievel. This is a large subset of DeepFashion, containing large pose and scale variations. It also has large diversities, large quantities, and rich annotations, including • 7,982 number of clothing items; • 52,712 number of in-shop clothes images, and ~200,000 cross-pose/scale pairs; • Each image is annotated by bounding box, clothing type and pose type.
15 © IBM Corporation, 2016 Gap envisions a future with augmented-reality 'dressing rooms' https://www.engadget.com/2017/01/30/gap-augmented-reality-dressing-rooms/
© IBM Corporation, 2016 Machine Learning 101 Presented by Chris Parsons Consultant – Artificial Intelligence IBM Systems Lab Services, UK and Ireland
IBM Systems | 17 • What is Machine Learning? • ML Use Cases by Industry • How to spot opportunities for Machine Learning in your business Agenda
18 © IBM Corporation, 2016 Hello, Machine Learning - MNIST
19 © IBM Corporation, 2016
20 © IBM Corporation, 2016
21 © IBM Corporation, 2016
22 © IBM Corporation, 2016
23 © IBM Corporation, 2016 Why has Machine Learning taken off?
24 © IBM Corporation, 2016 Simple Classification Problems For example, height and weight. The orange group represents children (low height/weight) and the blue adults.
25 © IBM Corporation, 2016 Simple Classification Problems For example, height and weight. The orange group represents children (low height/weight) and the blue adults.
26 © IBM Corporation, 2016 Cluster in a Cluster Time series data, complex sample surveys and longitudinal studies
27 © IBM Corporation, 2016 ML Use Cases By Industry
28 © IBM Corporation, 2016 Manufacturing Retail Healthcare & Life SciencesFinancial Services HospitalityUtilities • Predictive Maintenance • Process Optimisation • Demand Forecasting • Risk Analysis • Cross/Up Selling • Credit Checks • Customer Segmentation • Patient Triage • Proactive Health Management • Real Time Alerts and Diagnostics • Disease Identification • Inventory Planning • Cross-Channel marketing • Customer ROI and Lifetime Value • Smart Grid Management • Carbon Emissions • Customer Specific Pricing • Scheduling • Pricing • Social Media Analytics and Customer Sentiment
29 © IBM Corporation, 2016 90% Reduction in inspection times Significant Decrease in inspection times Significant Increase in checkable quantities/ day Significantly Decreased rate of Safety Risks • The utility provider inspects its vast transmission network via hand, with skilled workers placed into high-risk environments. This method is costly, occasionally dangerous, and difficult to scale. • To address this and augment worker productivity, the provider is seeking to deploy drones to make visual inspections of transmission towers. • To automate the image processing, the provider is using PowerAI to train a deep learning network to ID potential maintenance issues captured by the drones. • IBM is the only vendor who can provide the unique supremacy of NVIDIA Tesla P100 GPUs connected to POWER8 CPUs with NVIDIA NVLink technology for deep learning. • IBM’s integrated portfolio of solutions also allows the provider to not only apply deep learning but also in-memory DBMS and high speed storage to store and analyze various data using Power Systems and IBM ESS and Spectrum Scale. Asian Electric Utility Provider Maintenance Inspection
30 © IBM Corporation, 2016 90% Reduction in inspection times Improved Accuracy of Risk analysis in credit application process Increased Capital available for Investment and other revenue-generating opportunities Increased Responsiveness to clients. Accelerated Time to respond to clients. Applying inference in real time shortens the time to answer for all clients: improving the customer experience. • A major bank in Oceania is seeking to apply deep learning to the credit risk analysis for credit card applications. Their main goal from this undertaking is to explore options to add self-learning capabilities to the current credit risk marking process. • By using deep learning to improve the accuracy of risk analysis, the bank can determine how much capital needs to be held to cover that risk. • Even an improvement of just 1% accuracy in marking credit risk would reduce their capital holding requirements, allowing the freed up capital to be invested, generating more income for the bank and it’s account holders. • The solution uses the IBM S822LCs for HPC systems, each with four NVIDIA Tesla P100 GPUs and 1TB of memory, and the IBM Data Engine for NoSQL CAPI- attached flash system. • In addition, IBM is providing Apache Spark via IBM Spectrum Conductor for SQL and Spectrum Scale. Large Bank Credit Risk Analysis
31 © IBM Corporation, 2016
32 © IBM Corporation, 2016 Why Now? Explosion of data – all sources and availability of storage Processing capability and GPU’s Deep Learning frameworks
33 © IBM Corporation, 2016
34 © IBM Corporation, 2016
35 © IBM Corporation, 2016 Introducing PowerAI: Get Started Fast with Deep Learning Enabled by High Performance Computing Infrastructure Package of Pre-Compiled Major Deep Learning Frameworks Easy to install & get started with Deep Learning with Enterprise-Class Support Optimized for Performance To Take Advantage of NVLink
36 © IBM Corporation, 2016 PowerAI Platform Caffe NVCaffe TorchIBMCaffe DL4JTensorFlow OpenBLAS Theano Deep Learning Frameworks Accelerated Servers and Infrastructure for Scaling Spectrum Scale: High-Speed Parallel File System Scale to Cloud Cluster of NVLink Servers Coming Soon Bazel DIGITSNCCL Distributed Frameworks Supporting Libraries Chainer
37 © IBM Corporation, 2016 •“AI Vision,” a tool designed for developers with limited knowledge of deep learning to train and deploy deep learning models for computer vision. •Tools for data preparation: Integration with IBM Spectrum Conductor cluster virtualization software that integrates Apache Spark to ease the process of transforming unstructured as well as structured data sets to prepare them for deep learning training •Decreased training time: A distributed computing version of TensorFlow, a popular open-source machine learning framework first built by Google. This distributed version of TensorFlow takes advantage of a virtualized cluster of GPU-accelerated servers using cost-efficient, high-performance computing methods to bring deep learning training time down from weeks to hours •Easier model development: A new software tool called “DL Insight” that enables data scientists to rapidly get better accuracy from their deep learning models. This tool monitors the deep learning training process and automatically adjusts parameters for peak performance.
38 © IBM Corporation, 2016 “We are also working on a new solution that we call the Elinar GDPR AI Miner. Built on the IBM Power Systems and PowerAI platform combined with IBM BigInsights Text Analytics, it will use our unique AI capabilities to enable customers to mine huge amounts of GDPR data. Specifically, we will offer AI models for GDPR consent identification and data identification and extraction, which will help users to achieve compliance at lower cost and higher quality. These are just two of countless potential applications.” “The IBM Power S822LC server provides at least twice the performance of our x86 platform; everything is faster and easier. As a result, we can get new solutions to market very quickly, protecting our edge over the competition.” —Ari Juntunen, CTO, Elinar Oy Ltd
39 © IBM Corporation, 2016 Deep Learning – Example Industries Automotive and Transportation Security and Public Safety Consumer Web, Mobile, Retail Medicine and Biology Broadcast, Media and Entertainment • Autonomous driving: • Pedestrian detection • Accident avoidance Auto, trucking, heavy equipment, Tier 1 suppliers (Hyundai, Toyota, Komatsu, General Motors, Volvo) Titles: Director of Research, New Applications, “autonomous” in title • Video Surveillance • Image analysis • Facial recognition and detection Local and national police, public and private safety/ security (ADT, IViz, Pinkerton, Sentry) Titles: Head of Analytics • Image tagging • Speech recognition • Natural language • Sentiment analysis Hyperscale web companies, large retail (Google photos, Twitter, Woolworths, Aeon) Titles: VP/Dir Marketing, Chief Customer Officer, New Application Research • Drug discovery • Diagnostic assistance • Cancer cell detection Pharmaceutical, Medical equipment, Diagnostic labs (Takeda, Asian Pharma, Pfizer) Titles: Principal investigators, Dir of Scientific research • Captioning • Search • Recommendations • Real time translation Consumer facing companies with large streaming of existing media, or real time content Titles: VP/Dir of Marketing, Closed captioning roles, Dir Translation services
40 © IBM Corporation, 2016 How to spot opportunities for Machine Learning in your business
41 © IBM Corporation, 2016 Collecting Data What if you didn’t use a form? Images, Speech, Translation
42 © IBM Corporation, 2016 Improving Business Processes Talk to your teams If it’s easy, repeated work – you could probably automate it
43 © IBM Corporation, 2016 Preempt Your Users Make common tasks easy Save your users time Keep your users coming back
44 © IBM Corporation, 2016 Who to talk to?
45 © IBM Corporation, 2016 Innovation Leads Developer TeamsData Scientists • Looking for new ways to solve existing problems • Open to new ideas • Likely looking for ”quick wins” • Motivated by big differences for small cost • Great understanding of existing data • Looking for insight • Understand the doman • Hurt by long wait times for existing DL workloads • Don’t want to worry about the software/hardware stack • Probably being asked to look at ML/DL • Trying to integrate Machine Learning models into existing applications • Driven by ease of use • Performance nerds • Code portability a must
46 © IBM Corporation, 2016 Differences between the PowerAI frameworks
47 © IBM Corporation, 2016 TensorFlow Theano DIGITSCaffe More..Torch • Becoming De Facto standard • Easy to learn • Designed to support GPU execution • Awesome for Image Recognition (original use case) • Poor support for language modelling – due to legacy limitations • Web GUI • Easy to use • Abstracts features of other frameworks • Used @ Twitter and Facebook • Better debugging – Automatic Differentiation (reverse-mode) • No Compile Time https://github.com/zer0n/deepframeworks• Supported and Developed by “Worlds Largest” Academic DL Lab • Faster than TF and with a wider range of functions
48 © IBM Corporation, 2016 | 48 Introducing IBM Power System S822LC for HPC First Custom-Built GPU Accelerator Server with NVLink 2.5x Faster CPU-GPU Data Communication via NVLink NVLink 80 GB/s GPU P8 GPU GPU P8 GPU PCIe 32 GB/s GPU x86 GPU GPU x86 GPU No NVLink between CPU & GPU for x86 Servers: PCIe Bottleneck NVIDIA P100 Pascal GPU POWER8 NVLink Server x86 Servers with PCIe • Custom-built GPU Accelerator Server • High-Speed NVLink Connections between CPUs & GPUs and among GPUs • Features novel NVIDIA P100 Pascal GPU accelerator
© IBM Corporation, 2016 Questions? David Spurway – IBM Power Systems Product Manager Email: david.spurway@uk.ibm.com Phone: 07717 892 896 Twitter, LinkedIn, YouTube
50 © IBM Corporation, 2016 PowerAI Installation and Optimization Overview Designed to optimize deployments of PowerAI on Power System servers. This offering is appropriate both for those clients who are interested in exploring the capabilities and benefits of implementing PowerAI, as well as those clients who may currently have Deep Learning and HPC solutions implemented. Target Audience • Clients wanting to purse Deep Learning using PowerAI on Power Systems • Clients who have a 8335-GTB (Minsky) as part of the insertion program • Clients who have existing applications built on CAFFE, Torch, TensorFlow, Theano Benefits • The customer will have a working PowerAI proof of concept that they can use for further evaluation of PowerAI and Deep Learning based solutions • Skills transfer from our experts helps you fully exploit the capabilities PowerAI Qualifying Questions • Does the customer have an interest in Deep Learning/ Machine Learning? • Does the customer have a research team that needs assistance with setting up the infrastructure to take advantage of the NVIDIA Deep Learning capabilities that PowerAI enables • Has the account team offered a 8335-GTB (Minsky) as part of a larger Power transaction that could justify Pre-Sales or Post Sales funding for implementation Key Features • Design, Install and Configure PowerAI Ubuntu infrastructure (servers, network, file systems ) • Install CUDA Libraries & NVIDIA Drivers • Verify access to GPU and apply best practices • Install and validate PowerAI packages and sample applications utilizing such technologies such as Jupyter notebooks, and Spectrum Conductor • Install Spectrum Computer (LSF) job scheduling • BlueMind sofware installation and configuration (China only) • Provide Power infrastructure skills transfer to customer personnel • Assist in migration of existing customer ML-DL workload onto PowerAI • Optional Remote On-Demand Assistance as follow-up Duration The service varies depending on the size and complexity of the implementation, but can be customized to specific client requirements. Resources Learn more about PowerAI on IBM Power Systems at: https://www.ibm.com/us-en/marketplace/deep-learning-platform/ Team Contacts Owner Americas: Fred Robinson fdrobin@us.ibm.com Owner China: Yong Bao Tan tanyb@cn.ibm.com Owner Europe: David Uttley david_uttley@uk.ibm.com Owner India: Subramaniam Meenakshisundaram smeenaks@in.ibm.com Owner Japan: Shinichi Niimi nimi@jp.ibm.com Find a Lab Services Opportunity Manager in your area -> http://ibm.biz/LabServicesOM IBM Systems Lab Services — Power Systems
© IBM Corporation, 2016 Thank you! David Spurway – IBM Power Systems Product Manager Email: david.spurway@uk.ibm.com Phone: 07717 892 896 Twitter, LinkedIn, YouTube
52 © IBM Corporation, 2016 Trademarks and notes IBM Corporation 2015 • IBM, the IBM logo and ibm.com are registered trademarks, and other company, product, or service names may be trademarks or service marks of International Business Machines Corporation in the United States, other countries, or both. A current list of IBM trademarks is available on the web at “Copyright and trademark information” at www.ibm.com/legal/copytrade.shtml • Other company, product, and service names may be trademarks or service marks of others. • References in this publication to IBM products or services do not imply that IBM intends to make them available in all countries in which IBM operates. • IBM and IBM Credit LLC do not, nor intend to, offer or provide accounting, tax or legal advice to clients. Clients should consult with their own financial, tax and legal advisors. Any tax or accounting treatment decisions made by or on behalf of the client are the sole responsibility of the customer. • IBM Global Financing offerings are provided through IBM Credit LLC in the United States, IBM Canada Ltd. in Canada, and other IBM subsidiaries and divisions worldwide to qualified commercial and government clients. Rates and availability are based on a client’s credit rating, financing terms, offering type, equipment type and options, and may vary by country. Some offerings are not available in certain countries. Other restrictions may apply. Rates and offerings are subject to change, extension or withdrawal without notice.
53 © IBM Corporation, 2016 Special notices This document was developed for IBM offerings in the United States as of the date of publication. IBM may not make these offerings available in other countries, and the information is subject to change without notice. Consult your local IBM business contact for information on the IBM offerings available in your area. Information in this document concerning non-IBM products was obtained from the suppliers of these products or other public sources. Questions on the capabilities of non-IBM products should be addressed to the suppliers of those products. IBM may have patents or pending patent applications covering subject matter in this document. The furnishing of this document does not give you any license to these patents. Send license inquires, in writing, to IBM Director of Licensing, IBM Corporation, New Castle Drive, Armonk, NY 10504- 1785 USA. All statements regarding IBM future direction and intent are subject to change or withdrawal without notice, and represent goals and objectives only. The information contained in this document has not been submitted to any formal IBM test and is provided "AS IS" with no warranties or guarantees either expressed or implied. All examples cited or described in this document are presented as illustrations of the manner in which some IBM products can be used and the results that may be achieved. Actual environmental costs and performance characteristics will vary depending on individual client configurations and conditions. IBM Global Financing offerings are provided through IBM Credit Corporation in the United States and other IBM subsidiaries and divisions worldwide to qualified commercial and government clients. Rates are based on a client's credit rating, financing terms, offering type, equipment type and options, and may vary by country. Other restrictions may apply. Rates and offerings are subject to change, extension or withdrawal without notice. IBM is not responsible for printing errors in this document that result in pricing or information inaccuracies. All prices shown are IBM's United States suggested list prices and are subject to change without notice; reseller prices may vary. IBM hardware products are manufactured from new parts, or new and serviceable used parts. Regardless, our warranty terms apply. Any performance data contained in this document was determined in a controlled environment. Actual results may vary significantly and are dependent on many factors including system hardware configuration and software design and configuration. Some measurements quoted in this document may have been made on development-level systems. There is no guarantee these measurements will be the same on generally-available systems. Some measurements quoted in this document may have been estimated through extrapolation. Users of this document should verify the applicable data for their specific environment.

More Related Content

PPTX
Introduction to Machine Learning on IBM Power Systems
David Spurway
 
PPTX
IBM i and Linux case studies
David Spurway
 
PPTX
IBM Power Systems Outlook and Roadmap
David Spurway
 
PPTX
IBM Power Systems at the heart of Cognitive Solutions
David Spurway
 
PPTX
IBM Power Systems Update 2Q17
David Spurway
 
PDF
2016 August POWER Up Your Insights - IBM System Summit Mumbai
Anand Haridass
 
PDF
2016 Sept 1st - IBM Consultants & System Integrators Interchange - Big Data -...
Anand Haridass
 
PDF
Future of Power: Power Strategy and Offerings for Denmark - Steve Sibley
IBM Danmark
 
Introduction to Machine Learning on IBM Power Systems
David Spurway
 
IBM i and Linux case studies
David Spurway
 
IBM Power Systems Outlook and Roadmap
David Spurway
 
IBM Power Systems at the heart of Cognitive Solutions
David Spurway
 
IBM Power Systems Update 2Q17
David Spurway
 
2016 August POWER Up Your Insights - IBM System Summit Mumbai
Anand Haridass
 
2016 Sept 1st - IBM Consultants & System Integrators Interchange - Big Data -...
Anand Haridass
 
Future of Power: Power Strategy and Offerings for Denmark - Steve Sibley
IBM Danmark
 

What's hot (20)

PDF
IBM Storage for AI and Big Data
Tony Pearson
 
PDF
IBM Power Systems at FIS InFocus 2019
Paula Koziol
 
PDF
200 million qps on commodity hardware : Getting started with MySQL Cluster 7.4
Frazer Clement
 
PDF
ExecutiveEdge at Edge2015, Day 1 Recap
IBM Power Systems
 
PPTX
EMC Big Data Solutions Overview
walshe1
 
PDF
Emerson converged infrastructure (thermal-power-mgt.-security (gs)
Greg Stover
 
PPT
Why Infrastructure Matters for Big Data & Analytics
Rick Perret
 
PDF
Big Data: Infrastructure Implications for “The Enterprise of Things” - Stampe...
StampedeCon
 
PDF
Brochure : The EMC Big Data Solution
EMC
 
PPT
The client defined cloud final clementi
Mauricio Godoy
 
PDF
Success stories and recommendations from IBM clients
IBM India Smarter Computing
 
PDF
Martin Wildberger Presentation
Mauricio Godoy
 
PPTX
Netapp - An Agile Data Infrastructure to Power Your Cloud
Global Business Events
 
PDF
Yahoo & Hadoop
Mauricio Godoy
 
PPTX
Hadoop Boosts Profits in Media and Telecom Industry
DataWorks Summit
 
PPT
Steve Mills - Dispelling the Vapor Around Cloud Computing
Mauricio Godoy
 
PPT
IBM Cognos Business Intelligence using dashDB
IBM Cloud Data Services
 
PDF
ExecutiveEdge at Edge2015, Day 2 Recap
IBM Power Systems
 
PPT
InfoSphere Streams toolkits :Real-Time Analytics on Data in Motion
Avadhoot Patwardhan
 
PDF
IMS08 the momentum driving the ims future
Robert Hain
 
IBM Storage for AI and Big Data
Tony Pearson
 
IBM Power Systems at FIS InFocus 2019
Paula Koziol
 
200 million qps on commodity hardware : Getting started with MySQL Cluster 7.4
Frazer Clement
 
ExecutiveEdge at Edge2015, Day 1 Recap
IBM Power Systems
 
EMC Big Data Solutions Overview
walshe1
 
Emerson converged infrastructure (thermal-power-mgt.-security (gs)
Greg Stover
 
Why Infrastructure Matters for Big Data & Analytics
Rick Perret
 
Big Data: Infrastructure Implications for “The Enterprise of Things” - Stampe...
StampedeCon
 
Brochure : The EMC Big Data Solution
EMC
 
The client defined cloud final clementi
Mauricio Godoy
 
Success stories and recommendations from IBM clients
IBM India Smarter Computing
 
Martin Wildberger Presentation
Mauricio Godoy
 
Netapp - An Agile Data Infrastructure to Power Your Cloud
Global Business Events
 
Yahoo & Hadoop
Mauricio Godoy
 
Hadoop Boosts Profits in Media and Telecom Industry
DataWorks Summit
 
Steve Mills - Dispelling the Vapor Around Cloud Computing
Mauricio Godoy
 
IBM Cognos Business Intelligence using dashDB
IBM Cloud Data Services
 
ExecutiveEdge at Edge2015, Day 2 Recap
IBM Power Systems
 
InfoSphere Streams toolkits :Real-Time Analytics on Data in Motion
Avadhoot Patwardhan
 
IMS08 the momentum driving the ims future
Robert Hain
 
Ad

Similar to The world of Machine Learning, Deep Learning and PowerAI (20)

PDF
Big Data: Introducing BigInsights, IBM's Hadoop- and Spark-based analytical p...
Cynthia Saracco
 
PDF
Use cases for Hadoop and Big Data Analytics - InfoSphere BigInsights
Gord Sissons
 
PDF
OpenPOWER/POWER9 AI webinar
Ganesan Narayanasamy
 
PPTX
Building Confidence in Big Data - IBM Smarter Business 2013
IBM Sverige
 
PPT
Robert Lecklin - BigData is making a difference
IBM Sverige
 
PPTX
IBM Power Systems Update 1Q17
David Spurway
 
PPTX
Increase your ROI with Hadoop in Six Months - Presented by Dell, Cloudera and...
Cloudera, Inc.
 
PPTX
Cloud PoV
Noud van Mullekom
 
PDF
Has Your Data Gone Rogue?
Tony Pearson
 
PDF
UNLIMITED by Capgemini
Detlev Sandel
 
PDF
IBM Cloud Côte d'Azur Meetup - 20190328 - Optimisation
IBM France Lab
 
PPT
DevOps for Enterprise Systems - Rosalind Radcliffe
DevOps for Enterprise Systems
 
PPT
Ibm client strategy final [v4]
Morten Bjørklund
 
PDF
Fuel for the cognitive age: What's new in IBM predictive analytics
IBM SPSS Software
 
PDF
Cogntive computing ibm workshop Assirm15
Pietro Leo
 
PPTX
Demystifying Spark and Hadoop to gain Business Insights
Adrian Whitehead
 
PPTX
Move Cloud to the Core of your Business Strategy
ZNetLive
 
PPTX
Make from your it department a competitive differentiator for your business
Marcos Quezada
 
PPTX
IBM Power Systems Update 1Q17
David Spurway
 
PDF
The Data Lake: Empowering Your Data Science Team
Senturus
 
Big Data: Introducing BigInsights, IBM's Hadoop- and Spark-based analytical p...
Cynthia Saracco
 
Use cases for Hadoop and Big Data Analytics - InfoSphere BigInsights
Gord Sissons
 
OpenPOWER/POWER9 AI webinar
Ganesan Narayanasamy
 
Building Confidence in Big Data - IBM Smarter Business 2013
IBM Sverige
 
Robert Lecklin - BigData is making a difference
IBM Sverige
 
IBM Power Systems Update 1Q17
David Spurway
 
Increase your ROI with Hadoop in Six Months - Presented by Dell, Cloudera and...
Cloudera, Inc.
 
Cloud PoV
Noud van Mullekom
 
Has Your Data Gone Rogue?
Tony Pearson
 
UNLIMITED by Capgemini
Detlev Sandel
 
IBM Cloud Côte d'Azur Meetup - 20190328 - Optimisation
IBM France Lab
 
DevOps for Enterprise Systems - Rosalind Radcliffe
DevOps for Enterprise Systems
 
Ibm client strategy final [v4]
Morten Bjørklund
 
Fuel for the cognitive age: What's new in IBM predictive analytics
IBM SPSS Software
 
Cogntive computing ibm workshop Assirm15
Pietro Leo
 
Demystifying Spark and Hadoop to gain Business Insights
Adrian Whitehead
 
Move Cloud to the Core of your Business Strategy
ZNetLive
 
Make from your it department a competitive differentiator for your business
Marcos Quezada
 
IBM Power Systems Update 1Q17
David Spurway
 
The Data Lake: Empowering Your Data Science Team
Senturus
 
Ad

More from David Spurway (9)

PPTX
IBM i at the heart of Cognitive Solutions
David Spurway
 
PPTX
IBM Power Systems Announcement Update
David Spurway
 
PPTX
IBM i at the eart of cognitive solutions
David Spurway
 
PPTX
IBM Power Systems E850C and S824
David Spurway
 
PPTX
IBM Power for Life Sciences
David Spurway
 
PPTX
IBM Power Systems Update
David Spurway
 
PPTX
IBM Power Systems Open Innovation
David Spurway
 
PPTX
IBM Power Systems - enabling cloud solutions
David Spurway
 
PPTX
Announcement Overview 4Q14 (ext)
David Spurway
 
IBM i at the heart of Cognitive Solutions
David Spurway
 
IBM Power Systems Announcement Update
David Spurway
 
IBM i at the eart of cognitive solutions
David Spurway
 
IBM Power Systems E850C and S824
David Spurway
 
IBM Power for Life Sciences
David Spurway
 
IBM Power Systems Update
David Spurway
 
IBM Power Systems Open Innovation
David Spurway
 
IBM Power Systems - enabling cloud solutions
David Spurway
 
Announcement Overview 4Q14 (ext)
David Spurway
 

Recently uploaded (20)

PDF
Chapter 3 Spatial Domain Image Processing.pdf
Getnet Tigabie Askale -(GM)
 
PDF
Mobile App Security Testing_ A Comprehensive Guide.pdf
flufftailshop
 
PDF
[발표본] 너의 과제는 클라우드에 있어_KTDS_김동현_20250524.pdf
ssuserf8b8bd1
 
PDF
Advanced Soft Computing BINUS July 2025.pdf
University of Hertfordshire
 
DOCX
The AUB Centre for AI in Media Proposal.docx
GAONE9
 
PPTX
Big Data Technologies - Introduction.pptx
kuthubussaman1
 
PDF
Unlocking AI with Model Context Protocol (MCP)
Brian McKeiver
 
PPTX
Detection-First SIEM: Rule Types, Dashboards, and Threat-Informed Strategy
ReZa AdineH
 
PPTX
breach-and-attack-simulation-cybersecurity-india-chennai-defenderrabbit-2025....
defencerabbit Team
 
PPTX
Understanding_Digital_Forensics_Presentation.pptx
ImranKhan423233
 
PDF
Peak of Data & AI Encore- AI for Metadata and Smarter Workflows
Safe Software
 
PDF
GDG Cloud Iasi [PUBLIC] Florian Blaga - Unveiling the Evolution of Cybersecur...
athlonica
 
PDF
Spectral efficient network and resource selection model in 5G networks
IAESIJAI
 
PDF
Shreyas Phanse Resume: Experienced Backend Engineer | Java • Spring Boot • Ka...
SHREYAS PHANSE
 
PPTX
VMware vSphere Foundation How to Sell Presentation-Ver1.4-2-14-2024.pptx
blackmambaettijean
 
PDF
Dropbox Q2 2025 Financial Results & Investor Presentation
Dropbox
 
PDF
Build a system with the filesystem maintained by OSTree @ COSCUP 2025
Jian-Hong Pan
 
PPTX
20250228 LYD VKU AI Blended-Learning.pptx
DatVoNgoc
 
PDF
Review of recent advances in non-invasive hemoglobin estimation
IAESIJAI
 
PDF
Optimiser vos workloads AI/ML sur Amazon EC2 et AWS Graviton
Julien SIMON
 
Chapter 3 Spatial Domain Image Processing.pdf
Getnet Tigabie Askale -(GM)
 
Mobile App Security Testing_ A Comprehensive Guide.pdf
flufftailshop
 
[발표본] 너의 과제는 클라우드에 있어_KTDS_김동현_20250524.pdf
ssuserf8b8bd1
 
Advanced Soft Computing BINUS July 2025.pdf
University of Hertfordshire
 
The AUB Centre for AI in Media Proposal.docx
GAONE9
 
Big Data Technologies - Introduction.pptx
kuthubussaman1
 
Unlocking AI with Model Context Protocol (MCP)
Brian McKeiver
 
Detection-First SIEM: Rule Types, Dashboards, and Threat-Informed Strategy
ReZa AdineH
 
breach-and-attack-simulation-cybersecurity-india-chennai-defenderrabbit-2025....
defencerabbit Team
 
Understanding_Digital_Forensics_Presentation.pptx
ImranKhan423233
 
Peak of Data & AI Encore- AI for Metadata and Smarter Workflows
Safe Software
 
GDG Cloud Iasi [PUBLIC] Florian Blaga - Unveiling the Evolution of Cybersecur...
athlonica
 
Spectral efficient network and resource selection model in 5G networks
IAESIJAI
 
Shreyas Phanse Resume: Experienced Backend Engineer | Java • Spring Boot • Ka...
SHREYAS PHANSE
 
VMware vSphere Foundation How to Sell Presentation-Ver1.4-2-14-2024.pptx
blackmambaettijean
 
Dropbox Q2 2025 Financial Results & Investor Presentation
Dropbox
 
Build a system with the filesystem maintained by OSTree @ COSCUP 2025
Jian-Hong Pan
 
20250228 LYD VKU AI Blended-Learning.pptx
DatVoNgoc
 
Review of recent advances in non-invasive hemoglobin estimation
IAESIJAI
 
Optimiser vos workloads AI/ML sur Amazon EC2 et AWS Graviton
Julien SIMON
 

The world of Machine Learning, Deep Learning and PowerAI

  • 1. © IBM Corporation, 2016 The World of Machine Learning, Deep Learning and PowerAI – Deeper Dive Power Up Live with SCC – 6th June 2017 Presented by David Spurway IBM Power Systems Product Manager IBM Systems, UK and Ireland
  • 2. 2 © IBM Corporation, 2016 Augmented intelligence, Artificial Intelligence, Cognitive driving innovation Faster
  • 3. 3 © IBM Corporation, 2016 My friends at Uni…
  • 4. 4 © IBM Corporation, 2016 From Big Data to AI client journey
  • 5. 5 © IBM Corporation, 2016 OBSERVATION DECISIONINTERPRETATION EVALUATION 010101010101010111100010011001010111 0000000000010101010100000000000 111101011 11000 000000000000 111111 010101 101010 10101010100 Prescriptive Best Outcomes? Descriptive What Has Happened? Cognitive Learn Dynamically Predictive What Could Happen?
  • 6. 6 © IBM Corporation, 2016 010101010101010111100010011001010111 0000000000010101010100000000000 111101011 11000 000000000000 111111 010101 101010 10101010100 OBSERVATION DECISIONINTERPRETATION EVALUATION Prescriptive Best Outcomes? Descriptive What Has Happened? Cognitive Learn Dynamically Predictive What Could Happen? ACTIONDATA How many frauds during last month ? Per Country ? Which Transactions will be fraudulent ? What is the best action in light of potential fraud ? In Natural Language : « Explain me why this transaction is fraudulent ?
  • 7. 7 © IBM Corporation, 2016 Prescriptive Best Outcomes? Descriptive What Has Happened? Cognitive Learn Dynamically Predictive What Could Happen? - Artificial - Intelligence - Big Data - NLP Robot Knowledge Base Deep Learning Machine Learning010101010101010111100010011001010111 1000101 100010 1 1000101 111010111010 00000000000010101010100000000000 111101011
  • 8. 8 © IBM Corporation, 2016 Prepare the data ALL DATA Input VAR Training Data Test Data Machine Learning Algorithms Predictive Model PREDICTION ? Test Data ACCURACY ? Machine Learning Algorithms use training data to create a predictive model: its accuracy is tested on holdback data Machine Learning TensorFlow Caffee Torch Theano Chainer Spark ML …… Prepare Data Build/Train Model Deploy/Score Monitor/Refine Iterative Development
  • 9. 9 © IBM Corporation, 2016 My recent buyer’s journey…
  • 10. 10 © IBM Corporation, 2016 Some challenges… https://uk.pinterest.com/pin/197595502370835426/sent/?sender=5279 06524979412201&invite_code=7fc292525ac44aafa6736bdec95dd1b5 Speziale Floral Lace Fit & Flare Dress Items in this section are temporarily out of stock
  • 11. 11 © IBM Corporation, 2016 Where I ended up going… Petite Clothing Update your wardrobe with Wallis' stunning must have petite range. Designed for women who are 5'3" and under
  • 12. 12 © IBM Corporation, 2016 Deep Learning Goes to the Dogs • https://openpowerfoundation.org/blogs/deep-learning-goes-to-the-dogs/ • http://vision.stanford.edu/aditya86/ImageNetDogs/ • The Stanford Dogs dataset contains images of 120 breeds of dogs from around the world. This dataset has been built using images and annotation from ImageNet for the task of fine-grained image categorization. • https://www.youtube.com/watch?v=6ZRuTWpIo4M
  • 13. 13 © IBM Corporation, 2016 Example of Datasets available http://mmlab.ie.cuhk.edu.hk/projects/DeepFashion.html
  • 14. 14 © IBM Corporation, 2016 DeepFashion: In-shop Clothes Retrieval Details In-shop Clothes Retrieval Benchmark evaluates the performance of in- shop Clothes Retrievel. This is a large subset of DeepFashion, containing large pose and scale variations. It also has large diversities, large quantities, and rich annotations, including • 7,982 number of clothing items; • 52,712 number of in-shop clothes images, and ~200,000 cross-pose/scale pairs; • Each image is annotated by bounding box, clothing type and pose type.
  • 15. 15 © IBM Corporation, 2016 Gap envisions a future with augmented-reality 'dressing rooms' https://www.engadget.com/2017/01/30/gap-augmented-reality-dressing-rooms/
  • 16. © IBM Corporation, 2016 Machine Learning 101 Presented by Chris Parsons Consultant – Artificial Intelligence IBM Systems Lab Services, UK and Ireland
  • 17. IBM Systems | 17 • What is Machine Learning? • ML Use Cases by Industry • How to spot opportunities for Machine Learning in your business Agenda
  • 18. 18 © IBM Corporation, 2016 Hello, Machine Learning - MNIST
  • 19. 19 © IBM Corporation, 2016
  • 20. 20 © IBM Corporation, 2016
  • 21. 21 © IBM Corporation, 2016
  • 22. 22 © IBM Corporation, 2016
  • 23. 23 © IBM Corporation, 2016 Why has Machine Learning taken off?
  • 24. 24 © IBM Corporation, 2016 Simple Classification Problems For example, height and weight. The orange group represents children (low height/weight) and the blue adults.
  • 25. 25 © IBM Corporation, 2016 Simple Classification Problems For example, height and weight. The orange group represents children (low height/weight) and the blue adults.
  • 26. 26 © IBM Corporation, 2016 Cluster in a Cluster Time series data, complex sample surveys and longitudinal studies
  • 27. 27 © IBM Corporation, 2016 ML Use Cases By Industry
  • 28. 28 © IBM Corporation, 2016 Manufacturing Retail Healthcare & Life SciencesFinancial Services HospitalityUtilities • Predictive Maintenance • Process Optimisation • Demand Forecasting • Risk Analysis • Cross/Up Selling • Credit Checks • Customer Segmentation • Patient Triage • Proactive Health Management • Real Time Alerts and Diagnostics • Disease Identification • Inventory Planning • Cross-Channel marketing • Customer ROI and Lifetime Value • Smart Grid Management • Carbon Emissions • Customer Specific Pricing • Scheduling • Pricing • Social Media Analytics and Customer Sentiment
  • 29. 29 © IBM Corporation, 2016 90% Reduction in inspection times Significant Decrease in inspection times Significant Increase in checkable quantities/ day Significantly Decreased rate of Safety Risks • The utility provider inspects its vast transmission network via hand, with skilled workers placed into high-risk environments. This method is costly, occasionally dangerous, and difficult to scale. • To address this and augment worker productivity, the provider is seeking to deploy drones to make visual inspections of transmission towers. • To automate the image processing, the provider is using PowerAI to train a deep learning network to ID potential maintenance issues captured by the drones. • IBM is the only vendor who can provide the unique supremacy of NVIDIA Tesla P100 GPUs connected to POWER8 CPUs with NVIDIA NVLink technology for deep learning. • IBM’s integrated portfolio of solutions also allows the provider to not only apply deep learning but also in-memory DBMS and high speed storage to store and analyze various data using Power Systems and IBM ESS and Spectrum Scale. Asian Electric Utility Provider Maintenance Inspection
  • 30. 30 © IBM Corporation, 2016 90% Reduction in inspection times Improved Accuracy of Risk analysis in credit application process Increased Capital available for Investment and other revenue-generating opportunities Increased Responsiveness to clients. Accelerated Time to respond to clients. Applying inference in real time shortens the time to answer for all clients: improving the customer experience. • A major bank in Oceania is seeking to apply deep learning to the credit risk analysis for credit card applications. Their main goal from this undertaking is to explore options to add self-learning capabilities to the current credit risk marking process. • By using deep learning to improve the accuracy of risk analysis, the bank can determine how much capital needs to be held to cover that risk. • Even an improvement of just 1% accuracy in marking credit risk would reduce their capital holding requirements, allowing the freed up capital to be invested, generating more income for the bank and it’s account holders. • The solution uses the IBM S822LCs for HPC systems, each with four NVIDIA Tesla P100 GPUs and 1TB of memory, and the IBM Data Engine for NoSQL CAPI- attached flash system. • In addition, IBM is providing Apache Spark via IBM Spectrum Conductor for SQL and Spectrum Scale. Large Bank Credit Risk Analysis
  • 31. 31 © IBM Corporation, 2016
  • 32. 32 © IBM Corporation, 2016 Why Now? Explosion of data – all sources and availability of storage Processing capability and GPU’s Deep Learning frameworks
  • 33. 33 © IBM Corporation, 2016
  • 34. 34 © IBM Corporation, 2016
  • 35. 35 © IBM Corporation, 2016 Introducing PowerAI: Get Started Fast with Deep Learning Enabled by High Performance Computing Infrastructure Package of Pre-Compiled Major Deep Learning Frameworks Easy to install & get started with Deep Learning with Enterprise-Class Support Optimized for Performance To Take Advantage of NVLink
  • 36. 36 © IBM Corporation, 2016 PowerAI Platform Caffe NVCaffe TorchIBMCaffe DL4JTensorFlow OpenBLAS Theano Deep Learning Frameworks Accelerated Servers and Infrastructure for Scaling Spectrum Scale: High-Speed Parallel File System Scale to Cloud Cluster of NVLink Servers Coming Soon Bazel DIGITSNCCL Distributed Frameworks Supporting Libraries Chainer
  • 37. 37 © IBM Corporation, 2016 •“AI Vision,” a tool designed for developers with limited knowledge of deep learning to train and deploy deep learning models for computer vision. •Tools for data preparation: Integration with IBM Spectrum Conductor cluster virtualization software that integrates Apache Spark to ease the process of transforming unstructured as well as structured data sets to prepare them for deep learning training •Decreased training time: A distributed computing version of TensorFlow, a popular open-source machine learning framework first built by Google. This distributed version of TensorFlow takes advantage of a virtualized cluster of GPU-accelerated servers using cost-efficient, high-performance computing methods to bring deep learning training time down from weeks to hours •Easier model development: A new software tool called “DL Insight” that enables data scientists to rapidly get better accuracy from their deep learning models. This tool monitors the deep learning training process and automatically adjusts parameters for peak performance.
  • 38. 38 © IBM Corporation, 2016 “We are also working on a new solution that we call the Elinar GDPR AI Miner. Built on the IBM Power Systems and PowerAI platform combined with IBM BigInsights Text Analytics, it will use our unique AI capabilities to enable customers to mine huge amounts of GDPR data. Specifically, we will offer AI models for GDPR consent identification and data identification and extraction, which will help users to achieve compliance at lower cost and higher quality. These are just two of countless potential applications.” “The IBM Power S822LC server provides at least twice the performance of our x86 platform; everything is faster and easier. As a result, we can get new solutions to market very quickly, protecting our edge over the competition.” —Ari Juntunen, CTO, Elinar Oy Ltd
  • 39. 39 © IBM Corporation, 2016 Deep Learning – Example Industries Automotive and Transportation Security and Public Safety Consumer Web, Mobile, Retail Medicine and Biology Broadcast, Media and Entertainment • Autonomous driving: • Pedestrian detection • Accident avoidance Auto, trucking, heavy equipment, Tier 1 suppliers (Hyundai, Toyota, Komatsu, General Motors, Volvo) Titles: Director of Research, New Applications, “autonomous” in title • Video Surveillance • Image analysis • Facial recognition and detection Local and national police, public and private safety/ security (ADT, IViz, Pinkerton, Sentry) Titles: Head of Analytics • Image tagging • Speech recognition • Natural language • Sentiment analysis Hyperscale web companies, large retail (Google photos, Twitter, Woolworths, Aeon) Titles: VP/Dir Marketing, Chief Customer Officer, New Application Research • Drug discovery • Diagnostic assistance • Cancer cell detection Pharmaceutical, Medical equipment, Diagnostic labs (Takeda, Asian Pharma, Pfizer) Titles: Principal investigators, Dir of Scientific research • Captioning • Search • Recommendations • Real time translation Consumer facing companies with large streaming of existing media, or real time content Titles: VP/Dir of Marketing, Closed captioning roles, Dir Translation services
  • 40. 40 © IBM Corporation, 2016 How to spot opportunities for Machine Learning in your business
  • 41. 41 © IBM Corporation, 2016 Collecting Data What if you didn’t use a form? Images, Speech, Translation
  • 42. 42 © IBM Corporation, 2016 Improving Business Processes Talk to your teams If it’s easy, repeated work – you could probably automate it
  • 43. 43 © IBM Corporation, 2016 Preempt Your Users Make common tasks easy Save your users time Keep your users coming back
  • 44. 44 © IBM Corporation, 2016 Who to talk to?
  • 45. 45 © IBM Corporation, 2016 Innovation Leads Developer TeamsData Scientists • Looking for new ways to solve existing problems • Open to new ideas • Likely looking for ”quick wins” • Motivated by big differences for small cost • Great understanding of existing data • Looking for insight • Understand the doman • Hurt by long wait times for existing DL workloads • Don’t want to worry about the software/hardware stack • Probably being asked to look at ML/DL • Trying to integrate Machine Learning models into existing applications • Driven by ease of use • Performance nerds • Code portability a must
  • 46. 46 © IBM Corporation, 2016 Differences between the PowerAI frameworks
  • 47. 47 © IBM Corporation, 2016 TensorFlow Theano DIGITSCaffe More..Torch • Becoming De Facto standard • Easy to learn • Designed to support GPU execution • Awesome for Image Recognition (original use case) • Poor support for language modelling – due to legacy limitations • Web GUI • Easy to use • Abstracts features of other frameworks • Used @ Twitter and Facebook • Better debugging – Automatic Differentiation (reverse-mode) • No Compile Time https://github.com/zer0n/deepframeworks• Supported and Developed by “Worlds Largest” Academic DL Lab • Faster than TF and with a wider range of functions
  • 48. 48 © IBM Corporation, 2016 | 48 Introducing IBM Power System S822LC for HPC First Custom-Built GPU Accelerator Server with NVLink 2.5x Faster CPU-GPU Data Communication via NVLink NVLink 80 GB/s GPU P8 GPU GPU P8 GPU PCIe 32 GB/s GPU x86 GPU GPU x86 GPU No NVLink between CPU & GPU for x86 Servers: PCIe Bottleneck NVIDIA P100 Pascal GPU POWER8 NVLink Server x86 Servers with PCIe • Custom-built GPU Accelerator Server • High-Speed NVLink Connections between CPUs & GPUs and among GPUs • Features novel NVIDIA P100 Pascal GPU accelerator
  • 49. © IBM Corporation, 2016 Questions? David Spurway – IBM Power Systems Product Manager Email: david.spurway@uk.ibm.com Phone: 07717 892 896 Twitter, LinkedIn, YouTube
  • 50. 50 © IBM Corporation, 2016 PowerAI Installation and Optimization Overview Designed to optimize deployments of PowerAI on Power System servers. This offering is appropriate both for those clients who are interested in exploring the capabilities and benefits of implementing PowerAI, as well as those clients who may currently have Deep Learning and HPC solutions implemented. Target Audience • Clients wanting to purse Deep Learning using PowerAI on Power Systems • Clients who have a 8335-GTB (Minsky) as part of the insertion program • Clients who have existing applications built on CAFFE, Torch, TensorFlow, Theano Benefits • The customer will have a working PowerAI proof of concept that they can use for further evaluation of PowerAI and Deep Learning based solutions • Skills transfer from our experts helps you fully exploit the capabilities PowerAI Qualifying Questions • Does the customer have an interest in Deep Learning/ Machine Learning? • Does the customer have a research team that needs assistance with setting up the infrastructure to take advantage of the NVIDIA Deep Learning capabilities that PowerAI enables • Has the account team offered a 8335-GTB (Minsky) as part of a larger Power transaction that could justify Pre-Sales or Post Sales funding for implementation Key Features • Design, Install and Configure PowerAI Ubuntu infrastructure (servers, network, file systems ) • Install CUDA Libraries & NVIDIA Drivers • Verify access to GPU and apply best practices • Install and validate PowerAI packages and sample applications utilizing such technologies such as Jupyter notebooks, and Spectrum Conductor • Install Spectrum Computer (LSF) job scheduling • BlueMind sofware installation and configuration (China only) • Provide Power infrastructure skills transfer to customer personnel • Assist in migration of existing customer ML-DL workload onto PowerAI • Optional Remote On-Demand Assistance as follow-up Duration The service varies depending on the size and complexity of the implementation, but can be customized to specific client requirements. Resources Learn more about PowerAI on IBM Power Systems at: https://www.ibm.com/us-en/marketplace/deep-learning-platform/ Team Contacts Owner Americas: Fred Robinson fdrobin@us.ibm.com Owner China: Yong Bao Tan tanyb@cn.ibm.com Owner Europe: David Uttley david_uttley@uk.ibm.com Owner India: Subramaniam Meenakshisundaram smeenaks@in.ibm.com Owner Japan: Shinichi Niimi nimi@jp.ibm.com Find a Lab Services Opportunity Manager in your area -> http://ibm.biz/LabServicesOM IBM Systems Lab Services — Power Systems
  • 51. © IBM Corporation, 2016 Thank you! David Spurway – IBM Power Systems Product Manager Email: david.spurway@uk.ibm.com Phone: 07717 892 896 Twitter, LinkedIn, YouTube
  • 52. 52 © IBM Corporation, 2016 Trademarks and notes IBM Corporation 2015 • IBM, the IBM logo and ibm.com are registered trademarks, and other company, product, or service names may be trademarks or service marks of International Business Machines Corporation in the United States, other countries, or both. A current list of IBM trademarks is available on the web at “Copyright and trademark information” at www.ibm.com/legal/copytrade.shtml • Other company, product, and service names may be trademarks or service marks of others. • References in this publication to IBM products or services do not imply that IBM intends to make them available in all countries in which IBM operates. • IBM and IBM Credit LLC do not, nor intend to, offer or provide accounting, tax or legal advice to clients. Clients should consult with their own financial, tax and legal advisors. Any tax or accounting treatment decisions made by or on behalf of the client are the sole responsibility of the customer. • IBM Global Financing offerings are provided through IBM Credit LLC in the United States, IBM Canada Ltd. in Canada, and other IBM subsidiaries and divisions worldwide to qualified commercial and government clients. Rates and availability are based on a client’s credit rating, financing terms, offering type, equipment type and options, and may vary by country. Some offerings are not available in certain countries. Other restrictions may apply. Rates and offerings are subject to change, extension or withdrawal without notice.
  • 53. 53 © IBM Corporation, 2016 Special notices This document was developed for IBM offerings in the United States as of the date of publication. IBM may not make these offerings available in other countries, and the information is subject to change without notice. Consult your local IBM business contact for information on the IBM offerings available in your area. Information in this document concerning non-IBM products was obtained from the suppliers of these products or other public sources. Questions on the capabilities of non-IBM products should be addressed to the suppliers of those products. IBM may have patents or pending patent applications covering subject matter in this document. The furnishing of this document does not give you any license to these patents. Send license inquires, in writing, to IBM Director of Licensing, IBM Corporation, New Castle Drive, Armonk, NY 10504- 1785 USA. All statements regarding IBM future direction and intent are subject to change or withdrawal without notice, and represent goals and objectives only. The information contained in this document has not been submitted to any formal IBM test and is provided "AS IS" with no warranties or guarantees either expressed or implied. All examples cited or described in this document are presented as illustrations of the manner in which some IBM products can be used and the results that may be achieved. Actual environmental costs and performance characteristics will vary depending on individual client configurations and conditions. IBM Global Financing offerings are provided through IBM Credit Corporation in the United States and other IBM subsidiaries and divisions worldwide to qualified commercial and government clients. Rates are based on a client's credit rating, financing terms, offering type, equipment type and options, and may vary by country. Other restrictions may apply. Rates and offerings are subject to change, extension or withdrawal without notice. IBM is not responsible for printing errors in this document that result in pricing or information inaccuracies. All prices shown are IBM's United States suggested list prices and are subject to change without notice; reseller prices may vary. IBM hardware products are manufactured from new parts, or new and serviceable used parts. Regardless, our warranty terms apply. Any performance data contained in this document was determined in a controlled environment. Actual results may vary significantly and are dependent on many factors including system hardware configuration and software design and configuration. Some measurements quoted in this document may have been made on development-level systems. There is no guarantee these measurements will be the same on generally-available systems. Some measurements quoted in this document may have been estimated through extrapolation. Users of this document should verify the applicable data for their specific environment.

Editor's Notes

  • #20: Example handwriting, does everyone recognise these? How?
  • #21: Greyscale image, take each pixel value and flatten the matrix into a 1 dimensional array
  • #22: Let the computer decide what the weights it will apply to each pixel and what bias to use based on performance against known samples. Then perform a logistic regression (softmax) algorithm to the pixel values
  • #23: This is what the computer ”sees” red is negative weighting and blue is positive. So where it sees blue it’s expecting to see a character/higher greyscale value for that character. Then when it sees an unknown image it matches it based on these patterns.
  • #25: It’s easy for a human to classify these groups, and we could probably program a system with traditional logic to divide these two sets based on a line of best fit.
  • #26: It’s easy for a human to classify these groups, and we could probably program a system with traditional logic to divide these two sets based on a line of best fit.
  • #27: Now it’s a non trivial task to cliassify a new point based on this data, using traditional logic we could probably do it but it’s going to be a hideous codebase. We’re much better letting the system choose for us based on finding an ideal bias and weight.
  • #34: Infosys report 2017 (survey of 1600 people). AMPLIFYING HUMAN POTENTIAL 2 TOWARDS PURPOSEFUL ARTIFICIAL INTELLIGENCE
  • #36: PowerAI is a software toolkit with deep learning frameworks and building block software designed to run on IBM's highest performing server in its OpenPOWER LC line, the IBM Power S822LC for High Performance Computing, which features NVIDIA NVLink technology optimized for Power and NVIDIA's latest GPU technology
  • #40: Example organizations and targets Automotive and Transportation Global automotive, trucking, heavy equipment manufacturers (Toyota, Kia, Komatsu, General Motors, Volvo) Tier 1 suppliers of components to original equipment manufacturers Call on: Head of Research, New Applications, Anyone with “autonomous” in their title Security and Public Safety Local and national police, public and private security forces Corporate and institutional safety and security Call on: head of analytics Consumer web, mobile, retail Hyperscale and “close to hyperscale” web companies, gaming, organizations who’s business involves significant unstructured user data (example: google photos, twitter) Large retail, organizations focused on customer segmentation, analysis, and personalization Call on: marketing leadership, chief customer officer, new applications research Medicine and Biology Less defined segment Focus on computational biology, cross-discipline research Call on: principal investigators, head of scientific research Broadcast, media and entertainment Consumer facing companies with large streaming of existing media, or real time content Applications include live translation, user interaction (”second screen”), and personalization or recommendation (“If you liked this, then you might like…”) Call on: marketing leaders, anyone responsible for closed captioning or translation services Marketing executives. Better understanding of client perceptions through sentiment analysis, greater personalization and recommendations for ongoing client engagement IT Executives. IT Management is being asked to support new infrastructure for Deep Learning solutions, and to date the earliest providers have been 2nd and 3rd tier OEMs. IBM can provide enterprise level support, and a more efficient infrastructure: fewer servers for the same amount of work Developers, NVLink and the Tesla P100 GPU are changing how Deep Learning codes operate, allowing users to run more complex “training” workloads, allowing for better targeting and personalization.
  • #42: Everywhere your user has to input stuff from a form – what if you did it differently? Photos of your car instead of filling in that dull insurance form? Comparethemarket style
  • #43: Mundane repeated tasks – CRM application automatically take data from a call?
  • #44: Predict what your user is going to want to do next, and help them. Let users get back to the interesting work