Infuse Your Business with Machine Learning

Transcript

1. @craigsdickson Craig S. Dickson Customer Engineer Google Cloud
2. Infuse Your Business with Machine Learning
3. What is Machine Learning?
4. “Machine Learning. This is the next transformation … the programming paradigm is changing. Instead of programming a computer, you teach a computer to learn something and it does what you want.” Eric Schmidt Executive Chairman, Alphabet Inc.
5. Derive Predictive Insights from Data Data Algorithm Insight
6. Title here
7. Neural Networks - Are Made of Layers OUTPUT INPUT
8. … Are Trained with Labelled Examples labeled photos “cat” “dog” “car” “apple” “flower” OUTPUT
9. … And Can Then Be Used with Unlabeled Data unlabeled photo “cat”
10. Neural Networks Have Wide Applicability Input Output “rice” “restaurants in Seoul” “hello!” “A close up of a small child holding a stuffed animal.” 안녕하세요
11. How does Google use Machine Learning?
12. “We will move from mobile first to an AI first world. We see huge opportunities to dramatically improve how people work.” Sundar Pichai CEO, Google
13. Android Apps Gmail Maps Photos Speech Search Translation YouTube and many others ... Used across products: Deep Learning usage at Google
14. Search New York Giants San Francisco Giants
15. Street View Street name Street number Sign Business facade Sign Business name Traffic light Traffic sign Street number
16. Google Photos [glacier]
17. Google Translate
18. Smart Reply 10% of all responses sent on mobile
19. Meeting Nanny: Use images to take action
20. Machine Learning in Your Organisation
21. Common Use Cases by Industry • Predictive maintenance or condition monitoring • Warranty reserve estimation • Propensity to buy • Demand forecasting • Process optimization • Telematics Manufacturing • Predictive inventory planning • Recommendation engines • Upsell and cross-channel marketing • Market segmentation and targeting • Customer ROI and lifetime value Retail • Alerts and diagnostics from real-time patient data • Disease identification and risk satisfaction • Patient triage optimization • Proactive health management • Healthcare provider sentiment analysis Healthcare and Life Sciences • Aircraft scheduling • Dynamic pricing • Social media – consumer feedback and interaction analysis • Customer complaint resolution • Traffic patterns and congestion management Travel and Hospitality • Risk analytics and regulation • Customer Segmentation • Cross-selling and up-selling • Sales and marketing campaign management • Credit worthiness evaluation Financial Services • Power usage analytics • Seismic data processing • Carbon emissions and trading • Customer-specific pricing • Smart grid management • Energy demand and supply optimization Energy, Feedstock and Utilities
22. Machine Learning Use Cases Structured Data Classification/ Regression • Customer Churn Analysis • Product Diagnostics • Forecasting Recommendation • Content Personalization • Product X-Sells/Up-sells Anomaly Detection • Fraud Detection • Asset Sensor Diagnostics • Log Metric Anomalies Unstructured Data Image Analytics • Identify damaged shipments • Explicit Content Classification • Identify “styles” in images Text Analytics • Call Center log analysis • Language Identification • Topic Classification • Sentiment Analysis
23. Your Business Can Benefit From ML Simplify user input Adapt to user Infuse Your Apps with ML Anticipate users’ needs Creatively fulfill intent Delight Your Users Streamline your business processes Create a new line of business Fine-tune Your business
24. “Realistically a cloud product like GCP is the only way we can grow from an Australian-based company to a global company. By the end of 2018, we are looking to expand into Europe as well as Singapore, Canada and other countries. If we had the burden of looking to set up our own infrastructure, it simply wouldn’t be feasible.” —Dr Elliot Smith, Founder and CTO, Maxwell MRI HEALTHCARE IT STRATEGY / APPROACH Run the AI platform on infrastructure that could deliver massive scale in data storage and compute and could easily be accessed from any location. With GCP, Maxwell MRI has delivered results to clinicians in 10-15 minutes rather than up to a week and scaled to process 6,000 anonymised research cases, with another 144,000 forecast to be processed by the end of 2018. WHY GOOGLE Maxwell MRI chose GCP because it: • Provided highly scalable data storage and compute resources in a range of markets. • Enabled the business to focus on product development and customer service rather than infrastructure maintenance. • Delivered a flexible cost model that enabled the business to control its expenditure.
25. PLACEHOLDER: Salmat Use Case Slide
26. Machine Learning on Google Cloud Platform
27. 3 Keys to Successful Machine Learning Large Datasets Good Models Lots of Computation
28. Pre-trained ML Models Translation API Vision API Speech API Natural Language API Jobs API Video Intelligence API
29. Uses of the Google Vision API
30. Custom ML Models Machine Learning Engine TensorFlow
31. TensorFlow repository for “machine learning” category on GitHub #1 Sharing our tools with researchers and developers around the world
32. Machine Learning Data Pipeline TensorFlow BigQuery Cloud Pub/Sub Cloud Storage Cloud Datalab Cloud Dataprep Cloud Dataflow Cloud Machine Learning API Clients
33. Next Steps
34. Get Your Hands a Little Dirty GCP Free Tier • https://cloud.google.com/free/ Instructor Led Training Self Study Codelabs • TensorFlow for Poets ◦ https://goo.gl/bVjFIL • TensorFlow and Deep Learning, Without a PhD ◦ https://goo.gl/VSNhvw
35. We Can Help You Get Started Two day session that guides your business strategists towards formulating a concrete ML strategy that can be acted upon immediately. Discovery Workshop Long term immersion on Google campus working with our engineers to solve high impact business challenges through the power of machine learning Advanced Solutions Lab Technical advisory to assist with solution design and implementation of production ready workloads to Google Cloud Platform. Cloud Plan
36. Want To Go Deeper Today? Machine Learning at Scale - TensorFlow in the Cloud Win Woo & KJ Pittl 2:20pm in this room
37. Thank You. @craigsdickson