SlideShare a Scribd company logo
1 of 25
Download to read offline
RTB Optimizer: Behind the scenes witha Predictive API 
Nicolas KruchtenPAPIs.io –November 18, 2014 
REAL TIME MACHINE LEARNING 
DECISIONS AS A SERVICE
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
Real-Time Bidding for online advertising 
Real-Time 
Exchange 
Bidder 
Bidder 
Bidder 
Bidder 
Web 
Browser 
GET ad 
bid requests
Real-Time Bidding for online advertising 
Real-Time 
Exchange 
Bidder 
Bidder 
Bidder 
Bidder 
Web 
Browser 
ad 
bids 
auction
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
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
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
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
RTB Optimizer 
Bids API 
Outcomes API
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
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
Don’t buy lottery tickets! 
E[ value ] = payout * P( getting the payout )
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
RTB Optimizer 
Bids API 
E[ value ] 
Outcomes API 
P( outcome )
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
RTB Optimizer 
Probe 
Bids API 
E[ value ] 
Training 
Outcomes API 
P( outcome )
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
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
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%
RTB Optimizer 
Probe 
Bids API 
E[ value ] 
Training 
Outcomes API 
GLZ Classifier 
Stats Tables 
Real-Time 
Batch
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
Does it work? 
Classification success? ROC or calibration curves…
Does it work? 
Classification success? ROC and calibration curves… 
Optimization success? 80% reductions in cost-per-outcome…
Does it work? 
Classification success? ROC or calibration curves… 
Optimization success? 80% reductions in cost-per-outcome… 
Customer success! 25% monthly growth
Thanks! 
nicolas@datacratic.com 
REAL TIME MACHINE LEARNING 
DECISIONS AS A SERVICE

More Related Content

Similar to Nicolas Kruchten @ Datacratic

Mobile Ad Monetization for Games | Christian Calderon
Mobile Ad Monetization for Games | Christian CalderonMobile Ad Monetization for Games | Christian Calderon
Mobile Ad Monetization for Games | Christian CalderonJessica Tams
 
Szetela practcal ad words ai rocks
Szetela practcal ad words ai rocksSzetela practcal ad words ai rocks
Szetela practcal ad words ai rocksDavid Szetela
 
Artificial Intelligence and Machine Learning in PPC - David Szetela
Artificial Intelligence and Machine Learning in PPC - David SzetelaArtificial Intelligence and Machine Learning in PPC - David Szetela
Artificial Intelligence and Machine Learning in PPC - David SzetelaState of Search Conference
 
RTBMA ECIR 2016 tutorial
RTBMA ECIR 2016 tutorialRTBMA ECIR 2016 tutorial
RTBMA ECIR 2016 tutorialShuai Yuan
 
AI and Machine Language in PPC
AI and Machine Language in PPCAI and Machine Language in PPC
AI and Machine Language in PPCDavid Szetela
 
Response prediction for display advertising - WSDM 2014
Response prediction for display advertising - WSDM 2014Response prediction for display advertising - WSDM 2014
Response prediction for display advertising - WSDM 2014Olivier Chapelle
 
Vpon - 廣告效果導向為基礎的行動廣告系統
Vpon - 廣告效果導向為基礎的行動廣告系統Vpon - 廣告效果導向為基礎的行動廣告系統
Vpon - 廣告效果導向為基礎的行動廣告系統Vpon
 
Preffered account structure and performance tracking in Apple Search Ads
Preffered account structure and performance tracking in Apple Search AdsPreffered account structure and performance tracking in Apple Search Ads
Preffered account structure and performance tracking in Apple Search AdsAnna Yurchuk
 
Offer Platform
Offer PlatformOffer Platform
Offer Platform3Di
 
Data-Driven Reserve Prices for Social Advertising Auctions at LinkedIn
Data-Driven Reserve Prices for Social Advertising Auctions at LinkedInData-Driven Reserve Prices for Social Advertising Auctions at LinkedIn
Data-Driven Reserve Prices for Social Advertising Auctions at LinkedInKun Liu
 
Digital Marketing Course Week 8: Online Advertising
Digital Marketing Course Week 8: Online AdvertisingDigital Marketing Course Week 8: Online Advertising
Digital Marketing Course Week 8: Online AdvertisingAyca Turhan
 
apidays LIVE London 2021 - Productization of APIs by Jay Dembani, Avalara
apidays LIVE London 2021 - Productization of APIs by Jay Dembani, Avalaraapidays LIVE London 2021 - Productization of APIs by Jay Dembani, Avalara
apidays LIVE London 2021 - Productization of APIs by Jay Dembani, Avalaraapidays
 
SiteScout August Buyer Strategy Webinar
SiteScout August Buyer Strategy WebinarSiteScout August Buyer Strategy Webinar
SiteScout August Buyer Strategy Webinarsitescout
 
Making advertising personal, 4th NL Recommenders Meetup
Making advertising personal, 4th NL Recommenders MeetupMaking advertising personal, 4th NL Recommenders Meetup
Making advertising personal, 4th NL Recommenders MeetupOlivier Koch
 
How to buy traffic from Facebook, Instagram and Facebook Audience Network
How to buy traffic from Facebook, Instagram and Facebook Audience NetworkHow to buy traffic from Facebook, Instagram and Facebook Audience Network
How to buy traffic from Facebook, Instagram and Facebook Audience NetworkTravelpayouts
 
Maximize Your Revenue in Paid Search
Maximize Your Revenue in Paid SearchMaximize Your Revenue in Paid Search
Maximize Your Revenue in Paid SearchHanapin Marketing
 
Why is programmatic taking off? What is this revolution all about?
Why is programmatic taking off?  What is this revolution all about?Why is programmatic taking off?  What is this revolution all about?
Why is programmatic taking off? What is this revolution all about?Datacratic
 
Reel bid insightd-eproject
Reel bid   insightd-eprojectReel bid   insightd-eproject
Reel bid insightd-eprojectMudit Uppal
 

Similar to Nicolas Kruchten @ Datacratic (20)

Mobile Ad Monetization for Games | Christian Calderon
Mobile Ad Monetization for Games | Christian CalderonMobile Ad Monetization for Games | Christian Calderon
Mobile Ad Monetization for Games | Christian Calderon
 
Szetela practcal ad words ai rocks
Szetela practcal ad words ai rocksSzetela practcal ad words ai rocks
Szetela practcal ad words ai rocks
 
Artificial Intelligence and Machine Learning in PPC - David Szetela
Artificial Intelligence and Machine Learning in PPC - David SzetelaArtificial Intelligence and Machine Learning in PPC - David Szetela
Artificial Intelligence and Machine Learning in PPC - David Szetela
 
RTBMA ECIR 2016 tutorial
RTBMA ECIR 2016 tutorialRTBMA ECIR 2016 tutorial
RTBMA ECIR 2016 tutorial
 
AI and Machine Language in PPC
AI and Machine Language in PPCAI and Machine Language in PPC
AI and Machine Language in PPC
 
Response prediction for display advertising - WSDM 2014
Response prediction for display advertising - WSDM 2014Response prediction for display advertising - WSDM 2014
Response prediction for display advertising - WSDM 2014
 
RTB Bid Landscape in Adform
RTB Bid Landscape in AdformRTB Bid Landscape in Adform
RTB Bid Landscape in Adform
 
Vpon - 廣告效果導向為基礎的行動廣告系統
Vpon - 廣告效果導向為基礎的行動廣告系統Vpon - 廣告效果導向為基礎的行動廣告系統
Vpon - 廣告效果導向為基礎的行動廣告系統
 
Preffered account structure and performance tracking in Apple Search Ads
Preffered account structure and performance tracking in Apple Search AdsPreffered account structure and performance tracking in Apple Search Ads
Preffered account structure and performance tracking in Apple Search Ads
 
Offer Platform
Offer PlatformOffer Platform
Offer Platform
 
Smartone v1.0
Smartone v1.0Smartone v1.0
Smartone v1.0
 
Data-Driven Reserve Prices for Social Advertising Auctions at LinkedIn
Data-Driven Reserve Prices for Social Advertising Auctions at LinkedInData-Driven Reserve Prices for Social Advertising Auctions at LinkedIn
Data-Driven Reserve Prices for Social Advertising Auctions at LinkedIn
 
Digital Marketing Course Week 8: Online Advertising
Digital Marketing Course Week 8: Online AdvertisingDigital Marketing Course Week 8: Online Advertising
Digital Marketing Course Week 8: Online Advertising
 
apidays LIVE London 2021 - Productization of APIs by Jay Dembani, Avalara
apidays LIVE London 2021 - Productization of APIs by Jay Dembani, Avalaraapidays LIVE London 2021 - Productization of APIs by Jay Dembani, Avalara
apidays LIVE London 2021 - Productization of APIs by Jay Dembani, Avalara
 
SiteScout August Buyer Strategy Webinar
SiteScout August Buyer Strategy WebinarSiteScout August Buyer Strategy Webinar
SiteScout August Buyer Strategy Webinar
 
Making advertising personal, 4th NL Recommenders Meetup
Making advertising personal, 4th NL Recommenders MeetupMaking advertising personal, 4th NL Recommenders Meetup
Making advertising personal, 4th NL Recommenders Meetup
 
How to buy traffic from Facebook, Instagram and Facebook Audience Network
How to buy traffic from Facebook, Instagram and Facebook Audience NetworkHow to buy traffic from Facebook, Instagram and Facebook Audience Network
How to buy traffic from Facebook, Instagram and Facebook Audience Network
 
Maximize Your Revenue in Paid Search
Maximize Your Revenue in Paid SearchMaximize Your Revenue in Paid Search
Maximize Your Revenue in Paid Search
 
Why is programmatic taking off? What is this revolution all about?
Why is programmatic taking off?  What is this revolution all about?Why is programmatic taking off?  What is this revolution all about?
Why is programmatic taking off? What is this revolution all about?
 
Reel bid insightd-eproject
Reel bid   insightd-eprojectReel bid   insightd-eproject
Reel bid insightd-eproject
 

More from PAPIs.io

Shortening the time from analysis to deployment with ml as-a-service — Luiz A...
Shortening the time from analysis to deployment with ml as-a-service — Luiz A...Shortening the time from analysis to deployment with ml as-a-service — Luiz A...
Shortening the time from analysis to deployment with ml as-a-service — Luiz A...PAPIs.io
 
Feature engineering — HJ Van Veen (Nubank) @@PAPIs Connect — São Paulo 2017
Feature engineering — HJ Van Veen (Nubank) @@PAPIs Connect — São Paulo 2017Feature engineering — HJ Van Veen (Nubank) @@PAPIs Connect — São Paulo 2017
Feature engineering — HJ Van Veen (Nubank) @@PAPIs Connect — São Paulo 2017PAPIs.io
 
Extracting information from images using deep learning and transfer learning ...
Extracting information from images using deep learning and transfer learning ...Extracting information from images using deep learning and transfer learning ...
Extracting information from images using deep learning and transfer learning ...PAPIs.io
 
Discovering the hidden treasure of data using graph analytic — Ana Paula Appe...
Discovering the hidden treasure of data using graph analytic — Ana Paula Appe...Discovering the hidden treasure of data using graph analytic — Ana Paula Appe...
Discovering the hidden treasure of data using graph analytic — Ana Paula Appe...PAPIs.io
 
Deep learning for sentiment analysis — André Barbosa (elo7) @PAPIs Connect — ...
Deep learning for sentiment analysis — André Barbosa (elo7) @PAPIs Connect — ...Deep learning for sentiment analysis — André Barbosa (elo7) @PAPIs Connect — ...
Deep learning for sentiment analysis — André Barbosa (elo7) @PAPIs Connect — ...PAPIs.io
 
Building machine learning service in your business — Eric Chen (Uber) @PAPIs ...
Building machine learning service in your business — Eric Chen (Uber) @PAPIs ...Building machine learning service in your business — Eric Chen (Uber) @PAPIs ...
Building machine learning service in your business — Eric Chen (Uber) @PAPIs ...PAPIs.io
 
Building machine learning applications locally with Spark — Joel Pinho Lucas ...
Building machine learning applications locally with Spark — Joel Pinho Lucas ...Building machine learning applications locally with Spark — Joel Pinho Lucas ...
Building machine learning applications locally with Spark — Joel Pinho Lucas ...PAPIs.io
 
Battery log data mining — Ramon Oliveira (Datart) @PAPIs Connect — São Paulo ...
Battery log data mining — Ramon Oliveira (Datart) @PAPIs Connect — São Paulo ...Battery log data mining — Ramon Oliveira (Datart) @PAPIs Connect — São Paulo ...
Battery log data mining — Ramon Oliveira (Datart) @PAPIs Connect — São Paulo ...PAPIs.io
 
A tensorflow recommending system for news — Fabrício Vargas Matos (Hearst tv)...
A tensorflow recommending system for news — Fabrício Vargas Matos (Hearst tv)...A tensorflow recommending system for news — Fabrício Vargas Matos (Hearst tv)...
A tensorflow recommending system for news — Fabrício Vargas Matos (Hearst tv)...PAPIs.io
 
Scaling machine learning as a service at Uber — Li Erran Li at #papis2016
Scaling machine learning as a service at Uber — Li Erran Li at #papis2016Scaling machine learning as a service at Uber — Li Erran Li at #papis2016
Scaling machine learning as a service at Uber — Li Erran Li at #papis2016PAPIs.io
 
Real-world applications of AI - Daniel Hulme @ PAPIs Connect
Real-world applications of AI - Daniel Hulme @ PAPIs ConnectReal-world applications of AI - Daniel Hulme @ PAPIs Connect
Real-world applications of AI - Daniel Hulme @ PAPIs ConnectPAPIs.io
 
Past, Present and Future of AI: a Fascinating Journey - Ramon Lopez de Mantar...
Past, Present and Future of AI: a Fascinating Journey - Ramon Lopez de Mantar...Past, Present and Future of AI: a Fascinating Journey - Ramon Lopez de Mantar...
Past, Present and Future of AI: a Fascinating Journey - Ramon Lopez de Mantar...PAPIs.io
 
Revolutionizing Offline Retail Pricing & Promotions with ML - Daniel Guhl @ P...
Revolutionizing Offline Retail Pricing & Promotions with ML - Daniel Guhl @ P...Revolutionizing Offline Retail Pricing & Promotions with ML - Daniel Guhl @ P...
Revolutionizing Offline Retail Pricing & Promotions with ML - Daniel Guhl @ P...PAPIs.io
 
Demystifying Deep Learning - Roberto Paredes Palacios @ PAPIs Connect
Demystifying Deep Learning - Roberto Paredes Palacios @ PAPIs ConnectDemystifying Deep Learning - Roberto Paredes Palacios @ PAPIs Connect
Demystifying Deep Learning - Roberto Paredes Palacios @ PAPIs ConnectPAPIs.io
 
Predictive APIs: What about Banking? - Natalino Busa @ PAPIs Connect
Predictive APIs: What about Banking? - Natalino Busa @ PAPIs ConnectPredictive APIs: What about Banking? - Natalino Busa @ PAPIs Connect
Predictive APIs: What about Banking? - Natalino Busa @ PAPIs ConnectPAPIs.io
 
Microdecision making in financial services - Greg Lamp @ PAPIs Connect
Microdecision making in financial services - Greg Lamp @ PAPIs ConnectMicrodecision making in financial services - Greg Lamp @ PAPIs Connect
Microdecision making in financial services - Greg Lamp @ PAPIs ConnectPAPIs.io
 
Engineering the Future of Our Choice with General AI - JoEllen Lukavec Koeste...
Engineering the Future of Our Choice with General AI - JoEllen Lukavec Koeste...Engineering the Future of Our Choice with General AI - JoEllen Lukavec Koeste...
Engineering the Future of Our Choice with General AI - JoEllen Lukavec Koeste...PAPIs.io
 
Distributed deep learning with spark on AWS - Vincent Van Steenbergen @ PAPIs...
Distributed deep learning with spark on AWS - Vincent Van Steenbergen @ PAPIs...Distributed deep learning with spark on AWS - Vincent Van Steenbergen @ PAPIs...
Distributed deep learning with spark on AWS - Vincent Van Steenbergen @ PAPIs...PAPIs.io
 
How to predict the future of shopping - Ulrich Kerzel @ PAPIs Connect
How to predict the future of shopping - Ulrich Kerzel @ PAPIs ConnectHow to predict the future of shopping - Ulrich Kerzel @ PAPIs Connect
How to predict the future of shopping - Ulrich Kerzel @ PAPIs ConnectPAPIs.io
 
The emergent opportunity of Big Data for Social Good - Nuria Oliver @ PAPIs C...
The emergent opportunity of Big Data for Social Good - Nuria Oliver @ PAPIs C...The emergent opportunity of Big Data for Social Good - Nuria Oliver @ PAPIs C...
The emergent opportunity of Big Data for Social Good - Nuria Oliver @ PAPIs C...PAPIs.io
 

More from PAPIs.io (20)

Shortening the time from analysis to deployment with ml as-a-service — Luiz A...
Shortening the time from analysis to deployment with ml as-a-service — Luiz A...Shortening the time from analysis to deployment with ml as-a-service — Luiz A...
Shortening the time from analysis to deployment with ml as-a-service — Luiz A...
 
Feature engineering — HJ Van Veen (Nubank) @@PAPIs Connect — São Paulo 2017
Feature engineering — HJ Van Veen (Nubank) @@PAPIs Connect — São Paulo 2017Feature engineering — HJ Van Veen (Nubank) @@PAPIs Connect — São Paulo 2017
Feature engineering — HJ Van Veen (Nubank) @@PAPIs Connect — São Paulo 2017
 
Extracting information from images using deep learning and transfer learning ...
Extracting information from images using deep learning and transfer learning ...Extracting information from images using deep learning and transfer learning ...
Extracting information from images using deep learning and transfer learning ...
 
Discovering the hidden treasure of data using graph analytic — Ana Paula Appe...
Discovering the hidden treasure of data using graph analytic — Ana Paula Appe...Discovering the hidden treasure of data using graph analytic — Ana Paula Appe...
Discovering the hidden treasure of data using graph analytic — Ana Paula Appe...
 
Deep learning for sentiment analysis — André Barbosa (elo7) @PAPIs Connect — ...
Deep learning for sentiment analysis — André Barbosa (elo7) @PAPIs Connect — ...Deep learning for sentiment analysis — André Barbosa (elo7) @PAPIs Connect — ...
Deep learning for sentiment analysis — André Barbosa (elo7) @PAPIs Connect — ...
 
Building machine learning service in your business — Eric Chen (Uber) @PAPIs ...
Building machine learning service in your business — Eric Chen (Uber) @PAPIs ...Building machine learning service in your business — Eric Chen (Uber) @PAPIs ...
Building machine learning service in your business — Eric Chen (Uber) @PAPIs ...
 
Building machine learning applications locally with Spark — Joel Pinho Lucas ...
Building machine learning applications locally with Spark — Joel Pinho Lucas ...Building machine learning applications locally with Spark — Joel Pinho Lucas ...
Building machine learning applications locally with Spark — Joel Pinho Lucas ...
 
Battery log data mining — Ramon Oliveira (Datart) @PAPIs Connect — São Paulo ...
Battery log data mining — Ramon Oliveira (Datart) @PAPIs Connect — São Paulo ...Battery log data mining — Ramon Oliveira (Datart) @PAPIs Connect — São Paulo ...
Battery log data mining — Ramon Oliveira (Datart) @PAPIs Connect — São Paulo ...
 
A tensorflow recommending system for news — Fabrício Vargas Matos (Hearst tv)...
A tensorflow recommending system for news — Fabrício Vargas Matos (Hearst tv)...A tensorflow recommending system for news — Fabrício Vargas Matos (Hearst tv)...
A tensorflow recommending system for news — Fabrício Vargas Matos (Hearst tv)...
 
Scaling machine learning as a service at Uber — Li Erran Li at #papis2016
Scaling machine learning as a service at Uber — Li Erran Li at #papis2016Scaling machine learning as a service at Uber — Li Erran Li at #papis2016
Scaling machine learning as a service at Uber — Li Erran Li at #papis2016
 
Real-world applications of AI - Daniel Hulme @ PAPIs Connect
Real-world applications of AI - Daniel Hulme @ PAPIs ConnectReal-world applications of AI - Daniel Hulme @ PAPIs Connect
Real-world applications of AI - Daniel Hulme @ PAPIs Connect
 
Past, Present and Future of AI: a Fascinating Journey - Ramon Lopez de Mantar...
Past, Present and Future of AI: a Fascinating Journey - Ramon Lopez de Mantar...Past, Present and Future of AI: a Fascinating Journey - Ramon Lopez de Mantar...
Past, Present and Future of AI: a Fascinating Journey - Ramon Lopez de Mantar...
 
Revolutionizing Offline Retail Pricing & Promotions with ML - Daniel Guhl @ P...
Revolutionizing Offline Retail Pricing & Promotions with ML - Daniel Guhl @ P...Revolutionizing Offline Retail Pricing & Promotions with ML - Daniel Guhl @ P...
Revolutionizing Offline Retail Pricing & Promotions with ML - Daniel Guhl @ P...
 
Demystifying Deep Learning - Roberto Paredes Palacios @ PAPIs Connect
Demystifying Deep Learning - Roberto Paredes Palacios @ PAPIs ConnectDemystifying Deep Learning - Roberto Paredes Palacios @ PAPIs Connect
Demystifying Deep Learning - Roberto Paredes Palacios @ PAPIs Connect
 
Predictive APIs: What about Banking? - Natalino Busa @ PAPIs Connect
Predictive APIs: What about Banking? - Natalino Busa @ PAPIs ConnectPredictive APIs: What about Banking? - Natalino Busa @ PAPIs Connect
Predictive APIs: What about Banking? - Natalino Busa @ PAPIs Connect
 
Microdecision making in financial services - Greg Lamp @ PAPIs Connect
Microdecision making in financial services - Greg Lamp @ PAPIs ConnectMicrodecision making in financial services - Greg Lamp @ PAPIs Connect
Microdecision making in financial services - Greg Lamp @ PAPIs Connect
 
Engineering the Future of Our Choice with General AI - JoEllen Lukavec Koeste...
Engineering the Future of Our Choice with General AI - JoEllen Lukavec Koeste...Engineering the Future of Our Choice with General AI - JoEllen Lukavec Koeste...
Engineering the Future of Our Choice with General AI - JoEllen Lukavec Koeste...
 
Distributed deep learning with spark on AWS - Vincent Van Steenbergen @ PAPIs...
Distributed deep learning with spark on AWS - Vincent Van Steenbergen @ PAPIs...Distributed deep learning with spark on AWS - Vincent Van Steenbergen @ PAPIs...
Distributed deep learning with spark on AWS - Vincent Van Steenbergen @ PAPIs...
 
How to predict the future of shopping - Ulrich Kerzel @ PAPIs Connect
How to predict the future of shopping - Ulrich Kerzel @ PAPIs ConnectHow to predict the future of shopping - Ulrich Kerzel @ PAPIs Connect
How to predict the future of shopping - Ulrich Kerzel @ PAPIs Connect
 
The emergent opportunity of Big Data for Social Good - Nuria Oliver @ PAPIs C...
The emergent opportunity of Big Data for Social Good - Nuria Oliver @ PAPIs C...The emergent opportunity of Big Data for Social Good - Nuria Oliver @ PAPIs C...
The emergent opportunity of Big Data for Social Good - Nuria Oliver @ PAPIs C...
 

Recently uploaded

High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...
High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...
High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...soniya singh
 
Call Girls In Dwarka 9654467111 Escorts Service
Call Girls In Dwarka 9654467111 Escorts ServiceCall Girls In Dwarka 9654467111 Escorts Service
Call Girls In Dwarka 9654467111 Escorts ServiceSapana Sha
 
Call Girls in Defence Colony Delhi 💯Call Us 🔝8264348440🔝
Call Girls in Defence Colony Delhi 💯Call Us 🔝8264348440🔝Call Girls in Defence Colony Delhi 💯Call Us 🔝8264348440🔝
Call Girls in Defence Colony Delhi 💯Call Us 🔝8264348440🔝soniya singh
 
Customer Service Analytics - Make Sense of All Your Data.pptx
Customer Service Analytics - Make Sense of All Your Data.pptxCustomer Service Analytics - Make Sense of All Your Data.pptx
Customer Service Analytics - Make Sense of All Your Data.pptxEmmanuel Dauda
 
办理学位证中佛罗里达大学毕业证,UCF成绩单原版一比一
办理学位证中佛罗里达大学毕业证,UCF成绩单原版一比一办理学位证中佛罗里达大学毕业证,UCF成绩单原版一比一
办理学位证中佛罗里达大学毕业证,UCF成绩单原版一比一F sss
 
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /WhatsappsBeautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsappssapnasaifi408
 
IMA MSN - Medical Students Network (2).pptx
IMA MSN - Medical Students Network (2).pptxIMA MSN - Medical Students Network (2).pptx
IMA MSN - Medical Students Network (2).pptxdolaknnilon
 
20240419 - Measurecamp Amsterdam - SAM.pdf
20240419 - Measurecamp Amsterdam - SAM.pdf20240419 - Measurecamp Amsterdam - SAM.pdf
20240419 - Measurecamp Amsterdam - SAM.pdfHuman37
 
Easter Eggs From Star Wars and in cars 1 and 2
Easter Eggs From Star Wars and in cars 1 and 2Easter Eggs From Star Wars and in cars 1 and 2
Easter Eggs From Star Wars and in cars 1 and 217djon017
 
9654467111 Call Girls In Munirka Hotel And Home Service
9654467111 Call Girls In Munirka Hotel And Home Service9654467111 Call Girls In Munirka Hotel And Home Service
9654467111 Call Girls In Munirka Hotel And Home ServiceSapana Sha
 
RadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdfRadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdfgstagge
 
2006_GasProcessing_HB (1).pdf HYDROCARBON PROCESSING
2006_GasProcessing_HB (1).pdf HYDROCARBON PROCESSING2006_GasProcessing_HB (1).pdf HYDROCARBON PROCESSING
2006_GasProcessing_HB (1).pdf HYDROCARBON PROCESSINGmarianagonzalez07
 
Predictive Analysis for Loan Default Presentation : Data Analysis Project PPT
Predictive Analysis for Loan Default  Presentation : Data Analysis Project PPTPredictive Analysis for Loan Default  Presentation : Data Analysis Project PPT
Predictive Analysis for Loan Default Presentation : Data Analysis Project PPTBoston Institute of Analytics
 
Generative AI for Social Good at Open Data Science East 2024
Generative AI for Social Good at Open Data Science East 2024Generative AI for Social Good at Open Data Science East 2024
Generative AI for Social Good at Open Data Science East 2024Colleen Farrelly
 
EMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM TRACKING WITH GOOGLE ANALYTICS.pptx
EMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM  TRACKING WITH GOOGLE ANALYTICS.pptxEMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM  TRACKING WITH GOOGLE ANALYTICS.pptx
EMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM TRACKING WITH GOOGLE ANALYTICS.pptxthyngster
 
Multiple time frame trading analysis -brianshannon.pdf
Multiple time frame trading analysis -brianshannon.pdfMultiple time frame trading analysis -brianshannon.pdf
Multiple time frame trading analysis -brianshannon.pdfchwongval
 
Identifying Appropriate Test Statistics Involving Population Mean
Identifying Appropriate Test Statistics Involving Population MeanIdentifying Appropriate Test Statistics Involving Population Mean
Identifying Appropriate Test Statistics Involving Population MeanMYRABACSAFRA2
 
RABBIT: A CLI tool for identifying bots based on their GitHub events.
RABBIT: A CLI tool for identifying bots based on their GitHub events.RABBIT: A CLI tool for identifying bots based on their GitHub events.
RABBIT: A CLI tool for identifying bots based on their GitHub events.natarajan8993
 
Predicting Salary Using Data Science: A Comprehensive Analysis.pdf
Predicting Salary Using Data Science: A Comprehensive Analysis.pdfPredicting Salary Using Data Science: A Comprehensive Analysis.pdf
Predicting Salary Using Data Science: A Comprehensive Analysis.pdfBoston Institute of Analytics
 

Recently uploaded (20)

Deep Generative Learning for All - The Gen AI Hype (Spring 2024)
Deep Generative Learning for All - The Gen AI Hype (Spring 2024)Deep Generative Learning for All - The Gen AI Hype (Spring 2024)
Deep Generative Learning for All - The Gen AI Hype (Spring 2024)
 
High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...
High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...
High Class Call Girls Noida Sector 39 Aarushi 🔝8264348440🔝 Independent Escort...
 
Call Girls In Dwarka 9654467111 Escorts Service
Call Girls In Dwarka 9654467111 Escorts ServiceCall Girls In Dwarka 9654467111 Escorts Service
Call Girls In Dwarka 9654467111 Escorts Service
 
Call Girls in Defence Colony Delhi 💯Call Us 🔝8264348440🔝
Call Girls in Defence Colony Delhi 💯Call Us 🔝8264348440🔝Call Girls in Defence Colony Delhi 💯Call Us 🔝8264348440🔝
Call Girls in Defence Colony Delhi 💯Call Us 🔝8264348440🔝
 
Customer Service Analytics - Make Sense of All Your Data.pptx
Customer Service Analytics - Make Sense of All Your Data.pptxCustomer Service Analytics - Make Sense of All Your Data.pptx
Customer Service Analytics - Make Sense of All Your Data.pptx
 
办理学位证中佛罗里达大学毕业证,UCF成绩单原版一比一
办理学位证中佛罗里达大学毕业证,UCF成绩单原版一比一办理学位证中佛罗里达大学毕业证,UCF成绩单原版一比一
办理学位证中佛罗里达大学毕业证,UCF成绩单原版一比一
 
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /WhatsappsBeautiful Sapna Vip  Call Girls Hauz Khas 9711199012 Call /Whatsapps
Beautiful Sapna Vip Call Girls Hauz Khas 9711199012 Call /Whatsapps
 
IMA MSN - Medical Students Network (2).pptx
IMA MSN - Medical Students Network (2).pptxIMA MSN - Medical Students Network (2).pptx
IMA MSN - Medical Students Network (2).pptx
 
20240419 - Measurecamp Amsterdam - SAM.pdf
20240419 - Measurecamp Amsterdam - SAM.pdf20240419 - Measurecamp Amsterdam - SAM.pdf
20240419 - Measurecamp Amsterdam - SAM.pdf
 
Easter Eggs From Star Wars and in cars 1 and 2
Easter Eggs From Star Wars and in cars 1 and 2Easter Eggs From Star Wars and in cars 1 and 2
Easter Eggs From Star Wars and in cars 1 and 2
 
9654467111 Call Girls In Munirka Hotel And Home Service
9654467111 Call Girls In Munirka Hotel And Home Service9654467111 Call Girls In Munirka Hotel And Home Service
9654467111 Call Girls In Munirka Hotel And Home Service
 
RadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdfRadioAdProWritingCinderellabyButleri.pdf
RadioAdProWritingCinderellabyButleri.pdf
 
2006_GasProcessing_HB (1).pdf HYDROCARBON PROCESSING
2006_GasProcessing_HB (1).pdf HYDROCARBON PROCESSING2006_GasProcessing_HB (1).pdf HYDROCARBON PROCESSING
2006_GasProcessing_HB (1).pdf HYDROCARBON PROCESSING
 
Predictive Analysis for Loan Default Presentation : Data Analysis Project PPT
Predictive Analysis for Loan Default  Presentation : Data Analysis Project PPTPredictive Analysis for Loan Default  Presentation : Data Analysis Project PPT
Predictive Analysis for Loan Default Presentation : Data Analysis Project PPT
 
Generative AI for Social Good at Open Data Science East 2024
Generative AI for Social Good at Open Data Science East 2024Generative AI for Social Good at Open Data Science East 2024
Generative AI for Social Good at Open Data Science East 2024
 
EMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM TRACKING WITH GOOGLE ANALYTICS.pptx
EMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM  TRACKING WITH GOOGLE ANALYTICS.pptxEMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM  TRACKING WITH GOOGLE ANALYTICS.pptx
EMERCE - 2024 - AMSTERDAM - CROSS-PLATFORM TRACKING WITH GOOGLE ANALYTICS.pptx
 
Multiple time frame trading analysis -brianshannon.pdf
Multiple time frame trading analysis -brianshannon.pdfMultiple time frame trading analysis -brianshannon.pdf
Multiple time frame trading analysis -brianshannon.pdf
 
Identifying Appropriate Test Statistics Involving Population Mean
Identifying Appropriate Test Statistics Involving Population MeanIdentifying Appropriate Test Statistics Involving Population Mean
Identifying Appropriate Test Statistics Involving Population Mean
 
RABBIT: A CLI tool for identifying bots based on their GitHub events.
RABBIT: A CLI tool for identifying bots based on their GitHub events.RABBIT: A CLI tool for identifying bots based on their GitHub events.
RABBIT: A CLI tool for identifying bots based on their GitHub events.
 
Predicting Salary Using Data Science: A Comprehensive Analysis.pdf
Predicting Salary Using Data Science: A Comprehensive Analysis.pdfPredicting Salary Using Data Science: A Comprehensive Analysis.pdf
Predicting Salary Using Data Science: A Comprehensive Analysis.pdf
 

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! nicolas@datacratic.com REAL TIME MACHINE LEARNING DECISIONS AS A SERVICE