A few Challenges to Make Machine Learning Easy
2 likes983 views
The document discusses the challenges of making machine learning easy. It notes that while machine learning is a key technique for making data-driven decisions, developing smart applications, and building predictive analytics, it is also complex due to complicated tools that do not scale well, costly solutions, and a lack of experts. The talk presents BigML as a cloud-based service that aims to make machine learning simple by allowing users to create models with a few lines of code or via an easy-to-use web interface. It discusses the challenges BigML faces in achieving breadth and depth of techniques, supporting diverse users, maintaining simplicity, scaling to large amounts of data, measuring true impact, and determining fair pricing models.
A few Challenges to Make Machine Learning Easy
- 1. June 3rd, 2013BigML Inc, 2013 Challenges to Make Machine Learning Easy ACM San Francisco Bay Area Professional Chapter Francisco J Martin, Ph.D. BigML Co-founder & CEO eBay Whitman Campus
- 2. June 3rd, 2013BigML Inc, 2013 2 Expert: Published papers at KDD, ICML, NIPS, etc or developed own ML algorithms used at large scale. Sampling the Audience Aficionado: Understands pros/cons of different techniques and/or can tweak algorithms as needed. Newbie: Just taking Coursera ML class or reading an introductory book to ML. Absolute beginner: ML sounds like science fiction Practitioner: Very familiar with ML packages (Weka, Scikit, R, etc).
- 3. June 3rd, 2013BigML Inc, 2013 3 Data, data everywhere A special report on managing information Why make ML easy? In the age of data, Machine Learning is the key component to: ‣ make data-driven decisions ‣ develop smart applications ‣ build predictive analytics
- 4. June 3rd, 2013BigML Inc, 2013 However, Machine Learning is COMPLEX: ‣tools are complicated and do not scale well ‣solutions are costly ‣e x p e r t s w i t h i n d u s t r y experience are scarce 4 Why make ML easy? http://ttic.uchicago.edu/~samory/
- 5. June 3rd, 2013BigML Inc, 2013 5 Why make ML easy?
- 6. June 3rd, 2013BigML Inc, 2013 6 Why make ML easy?
- 7. April, 2013BigML Inc, 2013 7 BigML A cloud-based service that makes Machine Learning SIMPLE $ bigmler --train customer2012.csv --test new_customers.csv --objective churn >>> from bigml.api import BigML >>> api = BigML() >>> source = api.create_source("s3://bigml-public/csv/sales.csv") >>> dataset = api.create_dataset(source) >>> model = api.create_model(dataset) $ curl https://bigml.io/model?$BIGML_AUTH -X POST -H "content-type: application/json" -d '{"dataset": "dataset/50ca447b3b56356ae0000029"}'
- 8. June 3rd, 2013BigML Inc, 2013 8 Agenda BigML web-based interface (10-15 min) Questions (10-15 min) $ bigmler --train customer2012.csv --test new_customers.csv --objective churn >>> from bigml.api import BigML >>> api = BigML() >>> source = api.create_source("s3://bigml-public/csv/iris.csv") >>> dataset = api.create_dataset(source) >>> model = api.create_model(dataset) $ curl https://bigml.io/dataset?$BIGML_AUTH -X POST -H "content-type: application/json" -d '{"source": "source/50ca447b3b56356ae0000029"}' BigML API, API Bindings, BigMLer (5 min) Challenges (10-15 min) #1 Machine Learning Breadth and Depth #2 User Diversity #3 Simplicity #4 Scalability #5 Measuring Impact #6 Pricing
- 9. June 3rd, 2013BigML Inc, 2013 9 How it works
- 10. June 3rd, 2013BigML Inc, 2013 10 BigML Resources csv, arff, xls https, s3, azure, odata Sources local and remote Datasets Stream histograms Statistics Models Interactive Compoundable Random Decision Forests Actionable: exportable to rules, code, pmml Predictions Form-based Predictions Question by Question Local predictions Evaluations Classification Regression Comparison
- 11. June 3rd, 2013BigML Inc, 2013 11 BigML API
- 12. June 3rd, 2013BigML Inc, 2013 12 3,500+ users 35,000+ models BigML
- 13. June 3rd, 2013BigML Inc, 2013 13 FREE subscription? mail your username to: [email protected]
- 14. June 3rd, 2013BigML Inc, 2013 14 Challenges #1 Machine Learning breadth and depth #2 User Diversity #3 Simplicity #4 Scalability #5 Measuring Machine Learning Impact #6 Pricing
- 15. June 3rd, 2013BigML Inc, 2013 15 ...or you can deal with that! #1 Supervised learning #2 Unsupervised learning #3 Semi-supervised learning #4 Reinforcement learning #5 Learning to Learn #1 machine learning breadth and depth
- 16. June 3rd, 2013BigML Inc, 2013 16 ...or you can deal with that!#1 machine learning breadth and depth
- 17. June 3rd, 2013BigML Inc, 2013 17 Phrase a problem as an ML task The stages of an ML application Data Wrangling Feature Engineering Learn from Data Pre-evaluate Measure Impact
- 18. June 3rd, 2013BigML Inc, 2013 18 Problems Techniques Applications Classification Regression Clustering Density Estimation Manifold learning Active learning etc. Just solving a couple of problems and using a few techniques thousands of applications can be developed churn prevention, date matching, decision making, diagnostics, fraud detection, detecting tumors, detecting investment opportunities, human body pose estimation, pedestrian tracking, predictive analytics, recommendation systems, risk analysis, spam detection, etc #1 machine learning breadth and depth
- 19. June 3rd, 2013BigML Inc, 2013 19 Understanding the past Predicting the future Why Trees first?
- 20. June 3rd, 2013BigML Inc, 2013 20 Why Trees?
- 21. June 3rd, 2013BigML Inc, 2013 21 A Machine Learning application requires more tasks (that are even more important) than just learning from data. Just solving one problem more will enable a huge number of applications more. What problem(s) to tackle next and which techniques to use? #1 machine learning breadth and depth
- 22. June 3rd, 2013BigML Inc, 2013 22 Experts Aficionados Practitioner Newbies Absolute beginners #2 user diversity How to prioritize what to build next? More features for the expert or simplifying more for the newbies?
- 23. June 3rd, 2013BigML Inc, 2013 23 Time-to-productivity + + Expertise #2 user diversity
- 24. June 3rd, 2013BigML Inc, 2013 24 #2 user diversity MBs PBs MBs PBs Actual size Size Most users believe their data is much bigger than it really is
- 25. June 3rd, 2013BigML Inc, 2013 25 NumberofJobs + + Size of Job #2 user diversity
- 26. June 3rd, 2013BigML Inc, 2013 26 #3 simplicity
- 27. June 3rd, 2013BigML Inc, 2013 27 “Any fool can make something complicated. It takes a genius to make it simple.” ― Woody Guthrie #3 simplicity
- 28. June 3rd, 2013BigML Inc, 2013 28 ‣install ‣configure ‣use ‣train ‣understand ‣test ‣pre-evaluate ‣measure impact ‣deploy ‣scale ‣access programmatically (API) #3 simplicity Simple means much more than a easy-to-use interface
- 29. June 3rd, 2013BigML Inc, 2013 29 #4 scalability N CONCURRENT JOBS from 1 CUSTOMER 1 JOB from 1 USER N JOBS from M CUSTOMERS
- 30. June 3rd, 2013BigML Inc, 2013 30 Infrastructure
- 31. June 3rd, 2013BigML Inc, 2013 31 #5 measuring machine learning impact
- 32. June 3rd, 2013BigML Inc, 2013 32 Measuring “actual” impact is complex and goes beyond traditional performance evaluation. Imagine that an algorithm predicts that user Alice is going to buy a Magic Potion. ‣ But Magic Potions are out of stock. ‣ Should we blame ‣ the algorithm for the “false positive” prediction? ‣ the data scientist for not including that feature? ‣ operations for running out of stock on things that customers want to buy? #5 measuring machine learning impact
- 33. June 3rd, 2013BigML Inc, 2013 33 Kiri Wagstaff, Machine Learning that Matters, ICML, 2012 The stages of an ML research program Very inspirational!!!
- 34. June 3rd, 2013BigML Inc, 2013 34 Phrase a problem as an ML task Data Wrangling Learn from Data The stages of an ML application Feature Engineering Pre-evaluate Measure Impact !!!!!
- 35. June 3rd, 2013BigML Inc, 2013 35 #6 pricing
- 36. June 3rd, 2013BigML Inc, 2013 36 #6 pricing
- 37. June 3rd, 2013BigML Inc, 2013 37 Pre-pay-as-you-go
- 38. June 3rd, 2013BigML Inc, 2013 38 Subscriptions
- 39. June 3rd, 2013BigML Inc, 2013 39 ...or you can deal with that! BigML 1-click model You can deal with this... Machine Learning made easy?
- 40. June 3rd, 2013BigML Inc, 2013 40 BigML 1-click model You can deal with this... ...or you can deal with that! Machine Learning made easy?
- 41. June 3rd, 2013BigML Inc, 2013 41 Ease-of-use + + 2013 Machine Learning made easy?
- 42. June 3rd, 2013BigML Inc, 2013 42 Ease-of-use + + 2013 2014 2015 2016 2017 2018 Machine Learning made Easy!!!
- 43. June 3rd, 2013BigML Inc, 2013 43 Questions
- 44. June 3rd, 2013BigML Inc, 2013 44 Unknown Model f : X -> Y Example: ideal credit approval formula Models M Example: set of candidate credit approval formulas Learning from Data Learning Algorithm Based on Learning from Data by Y. Abu-Mostafa, M. Magdon-Ismail and H. Lin Final Model g ~ f Example: learned credit approval formula Training Examples (x1, l1), (x2, l2), ..., (xN, lN) Example: historical records of credit customers x1 xN labelf1 f2 fn