Nicolas Kruchten @ Datacratic
1 like•454 views
Datacratic provides a Real-Time Bidding (RTB) Optimizer predictive API that helps online advertisers maximize campaign goals like clicks, views, and conversions. The API uses machine learning to predict the probability of an outcome for each bid request and recommends optimal bids based on these predictions in real-time. It is able to learn from scratch based on incoming bid stream data and provide insights into what features drive performance for different campaigns.
![Second price auctions
•First Price Auctions
–You bid $1, I bid $2: I win, and I pay $2
•RTB uses Second Price Auctions
–You bid $1, I bid $2: I win, and I pay $1
•Optimal bid = E[ value ]
–Say it’s worth $2 to me
–I will never bid more than $2
–If I bid $1.50 and you bid $1.75: I’ve lost an opportunity for $0.25 surplus!
–I should always bid $2](https://image.slidesharecdn.com/b58u8cfruzjsmqlsmnq0-signature-f4a0831a458d6bbb78c09b1a397c3517fe8a2c82e9751f039f832a3be97b108f-poli-141208101338-conversion-gate01/85/Nicolas-Kruchten-Datacratic-11-320.jpg)
![Don’t buy lottery tickets!
E[ value ] = payout * P( getting the payout )](https://image.slidesharecdn.com/b58u8cfruzjsmqlsmnq0-signature-f4a0831a458d6bbb78c09b1a397c3517fe8a2c82e9751f039f832a3be97b108f-poli-141208101338-conversion-gate01/85/Nicolas-Kruchten-Datacratic-12-320.jpg)
![What’s it to you?
•If client gets paid $10,000 for 1,000 then payout = $10E[ value | bid-request ] = $10 * P( conversion | bid-request )
•What was an economics problem is now a prediction problem
•We need to calibrate to predict true probabilities](https://image.slidesharecdn.com/b58u8cfruzjsmqlsmnq0-signature-f4a0831a458d6bbb78c09b1a397c3517fe8a2c82e9751f039f832a3be97b108f-poli-141208101338-conversion-gate01/85/Nicolas-Kruchten-Datacratic-13-320.jpg)
![RTB Optimizer
Bids API
E[ value ]
Outcomes API
P( outcome )](https://image.slidesharecdn.com/b58u8cfruzjsmqlsmnq0-signature-f4a0831a458d6bbb78c09b1a397c3517fe8a2c82e9751f039f832a3be97b108f-poli-141208101338-conversion-gate01/85/Nicolas-Kruchten-Datacratic-14-320.jpg)
![Collecting the data
•To compute P( X | Y ) we need examples of Y’s with an X label
•RTB Optimizer uses mix of strategies to meet campaign goals
•Probe strategy bids randomly to collect data
•Optimized strategy bids with E[ value]
•Automatic training/retraining when API see enough examples](https://image.slidesharecdn.com/b58u8cfruzjsmqlsmnq0-signature-f4a0831a458d6bbb78c09b1a397c3517fe8a2c82e9751f039f832a3be97b108f-poli-141208101338-conversion-gate01/85/Nicolas-Kruchten-Datacratic-15-320.jpg)
![RTB Optimizer
Probe
Bids API
E[ value ]
Training
Outcomes API
P( outcome )](https://image.slidesharecdn.com/b58u8cfruzjsmqlsmnq0-signature-f4a0831a458d6bbb78c09b1a397c3517fe8a2c82e9751f039f832a3be97b108f-poli-141208101338-conversion-gate01/85/Nicolas-Kruchten-Datacratic-16-320.jpg)
![RTB Optimizer
Probe
Bids API
E[ value ]
Training
Outcomes API
GLZ Classifier
Stats Tables
Real-Time
Batch](https://image.slidesharecdn.com/b58u8cfruzjsmqlsmnq0-signature-f4a0831a458d6bbb78c09b1a397c3517fe8a2c82e9751f039f832a3be97b108f-poli-141208101338-conversion-gate01/85/Nicolas-Kruchten-Datacratic-20-320.jpg)
Nicolas Kruchten @ Datacratic
- 1. RTB Optimizer: Behind the scenes witha Predictive API Nicolas KruchtenPAPIs.io –November 18, 2014 REAL TIME MACHINE LEARNING DECISIONS AS A SERVICE
- 2. About Datacratic •Software company specializing in high performance systems andmachine learning •30 employees, founded in 2009, based in Montréal, Québec, Canada with an office in New York •3 Predictive APIs in market today •Building a Machine Learning Database to help others build Predictive APIs and Apps
- 3. Real-Time Bidding for online advertising Real-Time Exchange Bidder Bidder Bidder Bidder Web Browser GET ad bid requests
- 4. Real-Time Bidding for online advertising Real-Time Exchange Bidder Bidder Bidder Bidder Web Browser ad bids auction
- 5. Real-Time Bidding for online advertising Real-Time Exchange Bidder Bidder Bidder Bidder Web Browser This happens millions of times per second Bidders must respond within 100 milliseconds ad bids auction
- 6. Real-Time Bidding for online advertising Real-Time Exchange Bidder Bidder Bidder Bidder Web Browser RTB Optimizer enables bidders to achieve campaign goals ad bids auction
- 7. Campaign goals •Advertising campaignsare typically outcome-oriented –Clicks –Video views –Conversions: app installs, purchases, sign-ups •e.g. Ad network has sold someone 1,000 outcomes for $1,000 •e.g. Advertiser has $1,000 to get as many outcomes as possible •Essentially maximize profit or minimize cost-per-outcome
- 8. Datacratic’s RTB Optimizer •Client bidder relays bid-requests to API, API tells it how to bid •Handles 100,000 queries per second, for 100s of campaign •API says which campaign should bid and how much •API also needs outcomes in real-time and campaign goals
- 9. RTB Optimizer Bids API Outcomes API
- 10. A Predictive API that learns •Datacratichas no proprietary data set •API can learn from scratch from the bid-request stream what works for each campaign: –Contextual features: website, time of day, banner size and placement –User features: geo-location, browser, language, # of impressions shown –Customer-provided data: about the user, about the website •Provides insightsinto what features are driving performance •Can re-use learningsfrom previous campaigns
- 11. Second price auctions •First Price Auctions –You bid $1, I bid $2: I win, and I pay $2 •RTB uses Second Price Auctions –You bid $1, I bid $2: I win, and I pay $1 •Optimal bid = E[ value ] –Say it’s worth $2 to me –I will never bid more than $2 –If I bid $1.50 and you bid $1.75: I’ve lost an opportunity for $0.25 surplus! –I should always bid $2
- 12. Don’t buy lottery tickets! E[ value ] = payout * P( getting the payout )
- 13. What’s it to you? •If client gets paid $10,000 for 1,000 then payout = $10E[ value | bid-request ] = $10 * P( conversion | bid-request ) •What was an economics problem is now a prediction problem •We need to calibrate to predict true probabilities
- 14. RTB Optimizer Bids API E[ value ] Outcomes API P( outcome )
- 15. Collecting the data •To compute P( X | Y ) we need examples of Y’s with an X label •RTB Optimizer uses mix of strategies to meet campaign goals •Probe strategy bids randomly to collect data •Optimized strategy bids with E[ value] •Automatic training/retraining when API see enough examples
- 16. RTB Optimizer Probe Bids API E[ value ] Training Outcomes API P( outcome )
- 17. Bias control •Never stop the probe strategy •Always need control group for evaluation, retraining •Risk of filter bubbles: future models trained on previous output •Bid requests are randomly routed to probe, less often over time •Models automatically back-tested before deployment
- 18. How to learn in real-time •Classify using bagged generalized linear models •Generate non-linear features with statistics tables •Periodically retrain classifier •Continuously update stats tables
- 19. Statistics Table by example Table Bucket Impressions Outcomes Outcomes/Impressions 95%Confidence Lower Bound on Outcomes/Impressions Browser Chrome 5M 3k 0.060% 0.058% Firefox 3M 1k 0.033% 0.031% Website abc.com 4M 2k 0.050% 0.048% xyz.com 1k 10 1.000% 0.481%
- 20. RTB Optimizer Probe Bids API E[ value ] Training Outcomes API GLZ Classifier Stats Tables Real-Time Batch
- 21. Implementation details (are everything) •100k requests per second, 10 millisecond latency, running 24/7,1 trillion predictions to date •Distributed system, written in C++ 11 •AWS: data in S3, training runs on Amazon EC2 spot market •http://opensource.datacratic.com/ –RTBkit –JML –StarCluster
- 22. Does it work? Classification success? ROC or calibration curves…
- 23. Does it work? Classification success? ROC and calibration curves… Optimization success? 80% reductions in cost-per-outcome…
- 24. Does it work? Classification success? ROC or calibration curves… Optimization success? 80% reductions in cost-per-outcome… Customer success! 25% monthly growth
- 25. Thanks! [email protected] REAL TIME MACHINE LEARNING DECISIONS AS A SERVICE