Scaling_Machine_Learning_at_Holiday_Extras_-_MUC.pdf

Transcript

1. Scaling Machine Learning at Holiday Extras REBECCA VICKERY | DATA SCIENTIST @vickdata
2. Travel planning is time consuming Airport parking Airport hotels Airport lounges Travel insurance Holiday money Port products Car hire Airport transfers 582 minutes Over 46 days* Travel Planning *Facebook commissioned consumer research company GfK
3. Optimising consumer decision making Airport parking Airport hotels Airport lounges Travel insurance Holiday money Port products Car hire Airport transfers Less Hassle. More Holiday Trip recommendations
4. The machine learning process Historic Data (Input) (Input Algorithm Learns Mapping Predictions (Output)
5. Automated bidding Ad targeting Channel optimisation 1 Ad spend 2 Commercial 3 Customer Experience 4 Marketing Lots of other processes to optimise Automated pricing Allocation Revenue optimisation Automated call handling Personalised experiences Intelligent messaging Optimising send frequency
6. Machine learning needs to scale
7. Deploying machine learning is hard Scaling is even harder
8. Tools - Data Scientists Open source Lack Software Development expertise Mainly Python c Flaticon
9. Tools - Software Engineers Different tools Lack ML/Data expertise Mainly Javascript c Flaticon
10. Data science process The wrong kind of independence c Flaticon
11. People Small data science team Science + software experts are rare c Flaticon
12. Two types of deployment
13. Bespoke Solutions “Ideas are worth nothing unless executed”, Derek Sivers c Daniel Moyo
14. Unused Models Many models never make it to production “Ideas are worth nothing unless executed”, Derek Sivers
15. Time to model deployment Model development = days to weeks Model deployment = weeks to never! “Ideas are worth nothing unless executed”, Derek Sivers
16. The Google Way
17. c Flaticon __init__.py task.py setup.py model.py Model Package
18. Repeatable, Reusable Process __init__.py task.py setup.py model.py Model Package
20. Not Quite!
21. Collaborative Project
22. ML Proxy (bespoke ML microservice)
23. Monitoring - Model Performance
24. Monitoring - AI Platform Performance
25. AI Platform A technical solution but also a strategic solution c Google
26. The right kind of independence c flaticon Data Scientists can use preferred tools
27. The right kind of independence c flaticon Repeatable process for deployment of most models
28. Faster time to production c flaticon Fully Managed service
29. Faster time to production c flaticon Fast response from Google Cloud Support
30. Extensibility c flaticon.com Multiple models and versions
31. Customisable c flaticon.com Regularly release support for newer versions of tools
32. Time to model deployment Model development = days to weeks Model deployment = hours to days “Ideas are worth nothing unless executed”, Derek Sivers
33. Less Hassle. More Holiday Trip recommendations
34. Thank you @vickdata