beat google s bidder using ml
play

Beat Googles bidder using ML Dolead RD team Arnaud Fouchet DOLEAD - PowerPoint PPT Presentation

Nov 12, 2023 •412 likes •860 views

Beat Googles bidder using ML Dolead RD team Arnaud Fouchet DOLEAD @ Py.Paris NOV 2018 WWW.DOLEAD.COM Paid search basics 2 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM Paid search basics 3 DOLEAD @ Py.Paris - NOV 2018

  1. Beat Google’s bidder using ML Dolead RD team Arnaud Fouchet DOLEAD @ Py.Paris –NOV 2018 WWW.DOLEAD.COM

  2. Paid search basics 2 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  3. Paid search basics 3 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  4. Paid search basics Click 4 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  5. Paid search basics Click Visit Landing page 5 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  6. Paid search basics Click Visit Landing page Subscription (victory) Not interested (defeat) 6 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  7. Paid search basics Typed: “BTS a distance” Impression Will he click ? Click Will he subscribe ? Subscription New user 7 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  8. Paid search basics Typed: “BTS a distance” Impression Will he click ? Click Will he subscribe ? Subscription 100$ question : how much should I pay the ad ? New user 8 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  9. Paid search basics Typed: “BTS a distance” Impression Google’s problem Will he click ? Click Will he subscribe ? Subscription 100$ question : how much should I pay the ad ? New user 9 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  10. Paid search basics Typed: “BTS a distance” Impression Google’s problem Will he click ? Click Will he subscribe ? Subscription 10$ question : how much should I pay the click ? New user 10 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  11. Paid search basics How much should I pay the click ? 1. Find revenue by click 11 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  12. Paid search basics How much should I pay the click ? 1. Find revenue by click 2. Pay less than that 12 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  13. Paid search basics How much should I pay the click ? 1. Find revenue by click (using Machine learning: search query → revenue by click) 2. Pay less than that 13 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  14. Paid search basics Typed: “BTS a distance” Impression Google’s problem Will he click ? Click Google’s bidder Will he subscribe ? Subscription 10$ question : how much should I pay the click ? New user 14 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  15. Paid search basics Typed: “BTS a distance” Impression Google’s problem Will he click ? Click Google’s bidder Will he subscribe ? Subscription 1$ for 2 questions : Do I accept giving Google all power ? New user Can we do better ? 15 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  16. Paid search basics Typed: “BTS a distance” Impression Google’s problem Will he click ? Click Google’s bidder Will he convert ? Conversion 1$ for 2 questions : Do I accept giving Google all power ? New user Can we do better ? CPA: Cost Per Acquisition (Cost of a conversion) 16 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  17. Basic ML steps Identify learnable outcome (revenue by click) ● Transform input ( user’s search query ) into features (X) ● Identify relevant loss function ● Try and optimize models ● 17 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  18. Our ML research 18 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  19. Our ML research 19 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  20. Our ML research 20 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  21. Consensus Log RF GBDT ... Reg 21 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  22. Our ML research 22 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  23. Neural networks 23 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  24. Paid search basics Typed: “BTS a distance” Impression Google’s problem Will he click ? Click Will he convert ? Conversion 10$ question : how much should I pay the click ? New user 24 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  25. Our ML research 25 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  26. Paid search basics Typed: “BTS a distance” Impression Google’s problem Will he click ? Click Will he convert ? Conversion 10$ question : how much should I pay the click ? New user 26 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  27. Our ML research 27 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  28. Our ML research 28 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  29. Consensus RF GBDT NN ... Better 29 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  30. Consensus RF GBDT NN ... Better Weighted Consensus b x a x RF c x NN ... GBDT 30 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  31. Consensus RF GBDT NN ... Better Weighted Consensus b x a x RF c x NN ... GBDT Learned model f RF GBDT NN ... 31 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  32. Stacked models x GBDT NN RF ... x’ Learning problem on (x’, y) 32 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  33. Stacked models x ➝ data leakage problem GBDT NN RF ... Train x’ Learning problem on (x’, y) 33 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  34. Stacked models x ➝ data leakage problem DEVIL GBDT NN RF Train x’ Learning problem on (x’, y) 34 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  35. Measure Stacked Models Perf Train Train Campaign ... Campaign 1 Campaign 2 Predictions Predictions GBDT GBDT NN NN RF RF Test Test ... ... 35 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  36. Measure Stacked Models Perf Train Train Campaign ... Campaign 1 Campaign 2 Predictions Predictions GBDT GBDT NN NN RF RF Test Test ... ... Test ! Campaign T+ ... Train Train Campaign T+1 Campaign T+2 Predictions Predictions GBDT GBDT NN RF NN RF Test Test ... ... 36 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  37. Our ML research 37 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  38. Bidder comparison: context A/B test Experiment 1: 50% bids Google Campaign Experiment 2: 50% bids Dolead 38 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  39. Bidder comparison: context A/B test 50 campaigns 4 weeks Goal : maximize conversions while respecting target CPA Experiment 1: 50% bids Google Campaign Experiment 2: 50% bids Dolead 39 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  40. Bidder comparison: % Below target CPA Raw results : - Google 33% - Dolead 63% Distribution CPA(Dolead) / target CPA(Google) / target CPA CPA 40 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  41. Bidder comparison: Conversions & CPA Conversions : In 67% of campaigns, Google had more convs Distribution Convs(Google) / Convs(Dolead) 41 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  42. Bidder comparison: Conversions & CPA Conversions : CPA : In 67% of campaigns, Google had more convs In 83% of campaigns, Dolead was cheaper Distribution Distribution Convs(Google) / Convs(Dolead) CPA(Google) / CPA(Dolead) 42 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  43. Bidder comparison: Conversions & CPA Conversions : CPA : In 67% of campaigns, Google had more convs In 83% of campaigns, Dolead was cheaper Distribution Distribution Google’s margin: 9.8K Dolead’s margin: 13.3K (~+35%) Convs(Google) / Convs(Dolead) CPA(Dolead) / CPA(Google) 43 DOLEAD @ Py.Paris - NOV 2018 WWW.DOLEAD.COM

  44. Thanks for your attention ! PyApero 13 décember 2018 www.pyapero.com DOLEAD @ Py.Paris –NOV 2018 WWW.DOLEAD.COM

Download Presentation
Download Policy: The content available on the website is offered to you 'AS IS' for your personal information and use only. It cannot be commercialized, licensed, or distributed on other websites without prior consent from the author. To download a presentation, simply click this link. If you encounter any difficulties during the download process, it's possible that the publisher has removed the file from their server.

Recommend

Beat by Beat: Classifying Cardiac Arrhythmias with Recurrent Neural Networks Patrick Schwab,

Beat by Beat: Classifying Cardiac Arrhythmias with Recurrent Neural Networks Patrick Schwab,

Normal Beat Atrial Fibrillation Other Arrhythmia Beat by Beat: Classifying Cardiac Arrhythmias with Recurrent Neural Networks Patrick Schwab, Gaetano C. Scebba, Jia Zhang, Marco Delai and Walter Karlen Mobile Health Systems Lab Institute for

603 views • 20 slides
GRASS VALLEY CHP Beat 31 Beat 31 is SR-49 starting at the Bear River (Nevada / Placer County

GRASS VALLEY CHP Beat 31 Beat 31 is SR-49 starting at the Bear River (Nevada / Placer County

GRASS VALLEY CHP Beat 31 Beat 31 is SR-49 starting at the Bear River (Nevada / Placer County line) to the south and extending north to mile post marker 13.16 at the beginning of the Golden Center Freeway Beat 31 2016 A Primary Collision

382 views • 9 slides
i PETE Efaw 42 set 2 2 pi for both items E of 3 I PrN fa setp.nu 24 Pr buys 3 I Ey Eber

i PETE Efaw 42 set 2 2 pi for both items E of 3 I PrN fa setp.nu 24 Pr buys 3 I Ey Eber

Maximization Revenue bidders I seller in F Vin bidder i's value Revenntaximizing truthful auction Vin Fp independently alll Thin If regular Vct Vickrey is anion then ten maximizing to any single parameter setting Hey bidder

245 views • 11 slides
Beat the Street Torbay 19 th September 31 st October Beat the Street turns your whole area

Beat the Street Torbay 19 th September 31 st October Beat the Street turns your whole area

Beat the Street Torbay 19 th September 31 st October Beat the Street turns your whole area into a walking and cycling game for the next 6 weeks! Everyone can play! Parents, grandparents, carers, friends neighbours, brothers,

386 views • 17 slides
arXiv:1706.03762v5 [cs.CL] 6 Dec 2017 Llion Jones Aidan N. Gomez ukasz Kaiser

arXiv:1706.03762v5 [cs.CL] 6 Dec 2017 Llion Jones Aidan N. Gomez ukasz Kaiser

Attention Is All You Need Ashish Vaswani Noam Shazeer Niki Parmar Jakob Uszkoreit Google Brain Google Brain Google Research Google Research avaswani@google.com noam@google.com nikip@google.com usz@google.com

295 views • 15 slides
The Biopharmaceutical Industrys Efforts to Beat Coronavirus Sharon Lamberton, MS, RN (State

The Biopharmaceutical Industrys Efforts to Beat Coronavirus Sharon Lamberton, MS, RN (State

The Biopharmaceutical Industrys Efforts to Beat Coronavirus Sharon Lamberton, MS, RN (State Policy) Mary Kosinski (Science & Regulatory Advocacy ) May 2020 Our Commitment to Beat Coronavirus 2 Our Commitment to Beat Coronavirus We are

274 views • 26 slides
What is Jungle Beat? Remember when kids were kids and

What is Jungle Beat? Remember when kids were kids and

What is Jungle Beat? Remember when kids were kids and playing meant grass burns, bare feet and muddy faces? Jungle Beat is a fun,

722 views • 39 slides
Google AdWords & Google Analytics Jenn Davidson What are they? Several different Google

Google AdWords & Google Analytics Jenn Davidson What are they? Several different Google

Google AdWords & Google Analytics Jenn Davidson What are they? Several different Google Ad certification exams. Sales, shopping, mobile sites etc. Google Ads- online advertising service where advertisers pay to display brief

706 views • 9 slides
Google Analytics Overview Whats Google Analytics? The Google Analytics

Google Analytics Overview Whats Google Analytics? The Google Analytics

Google Analytics Overview Whats Google Analytics? The Google Analytics implementation formula. Steps involved in installing Google Analytics. Tracking events, campaigns and ecommerce data. Creating goals and funnels

652 views • 40 slides
ROBOD: a Real-time Online Beat and Offbeat Drummer ock 1 , Florian Krebs 1 , 2 , Amaury Durand 3 ,

ROBOD: a Real-time Online Beat and Offbeat Drummer ock 1 , Florian Krebs 1 , 2 , Amaury Durand 3 ,

Beat tracking ROBOD Beat tracking Pattern tracking Conclusion ROBOD: a Real-time Online Beat and Offbeat Drummer ock 1 , Florian Krebs 1 , 2 , Amaury Durand 3 , Sebastian P oll 1 , Raminta Sebastian B Balsyte 1 1Department of

670 views • 14 slides
Economic Value of Google Hal Varian Chief Economist Google Value of Google What I'm not

Economic Value of Google Hal Varian Chief Economist Google Value of Google What I'm not

Economic Value of Google Hal Varian Chief Economist Google Value of Google What I'm not going to do Counterfactual estimate of world without Google Alternative histories are like playing tennis with the net down What I am going

681 views • 23 slides
RPC Metrics at Google JBD, Google (@rakyll) gRPC Metrics at Google JBD, Google (@rakyll)

RPC Metrics at Google JBD, Google (@rakyll) gRPC Metrics at Google JBD, Google (@rakyll)

RPC Metrics at Google JBD, Google (@rakyll) gRPC Metrics at Google JBD, Google (@rakyll) Request Metrics at Google JBD, Google (@rakyll) "100% is the wrong reliability target for basically everything." -- Benjamin Treynor

452 views • 20 slides
Testing BEAT 3.0 Structures, scenarios and results Henrik Nygrd, Vivi Fleming-Lehtinen,

Testing BEAT 3.0 Structures, scenarios and results Henrik Nygrd, Vivi Fleming-Lehtinen,

Testing BEAT 3.0 Structures, scenarios and results Henrik Nygrd, Vivi Fleming-Lehtinen, Ciarn Murray, Samuli Korpinen BalticBOOST Biodiv WS2 2016, 14-15.9.2016, Copenhagen BEAT 3.0 inputs used Spatial assessment units Hierarchical

581 views • 17 slides
WorkOne Indy Service Provision Bidder's Conference February 9, 2018 Agenda Welcome and

WorkOne Indy Service Provision Bidder's Conference February 9, 2018 Agenda Welcome and

WorkOne Indy Service Provision Bidder's Conference February 9, 2018 Agenda Welcome and introductions System Overview Purpose of RFP Proposal requirements RFP timeline Technical assistance EmployIndy guides the local workforce ecosystem

417 views • 28 slides
BID NO (PIC003/2020): REQUEST FOR PROPOSAL TO APPOINT A SUITABLY QUALIFIED BIDDER FOR THE

BID NO (PIC003/2020): REQUEST FOR PROPOSAL TO APPOINT A SUITABLY QUALIFIED BIDDER FOR THE

BID NO (PIC003/2020): REQUEST FOR PROPOSAL TO APPOINT A SUITABLY QUALIFIED BIDDER FOR THE PROVISION OF A LISTED INVESTMENT MANAGEMENT SOLUTION AND SUPPORT OF THE SOLUTION FOR A PERIOD OF TEN (10) YEARS AGENDA o WELCOME o ATTENDANCE REGISTER o

407 views • 13 slides
Leprosy with your help Lepra can beat it! India, Bangladesh and Mozambique Where Lepra works

Leprosy with your help Lepra can beat it! India, Bangladesh and Mozambique Where Lepra works

Leprosy with your help Lepra can beat it! India, Bangladesh and Mozambique Where Lepra works Think about this morning What did you do? www.beat-leprosy.org.uk What about this family ? No cosy beds No breakfast No

616 views • 18 slides
A Scalable and Nearly Uniform Generator of SAT Witnesses Supratik Chakraborty 1 , Kuldeep S Meel 2

A Scalable and Nearly Uniform Generator of SAT Witnesses Supratik Chakraborty 1 , Kuldeep S Meel 2

A Scalable and Nearly Uniform Generator of SAT Witnesses Supratik Chakraborty 1 , Kuldeep S Meel 2 , Moshe Y Vardi 2 1 Indian Institute of Technology Bombay, India 2 Department of Computer Science, Rice University CAV 2013 Life in the 21 st

469 views • 33 slides
Lecture 3.2: Cosets Matthew Macauley Department of Mathematical Sciences Clemson University

Lecture 3.2: Cosets Matthew Macauley Department of Mathematical Sciences Clemson University

Lecture 3.2: Cosets Matthew Macauley Department of Mathematical Sciences Clemson University http://www.math.clemson.edu/~macaule/ Math 4120, Modern Algebra M. Macauley (Clemson) Lecture 3.2: Cosets Math 4120, Modern Algebra 1 / 11 Overview

702 views • 11 slides
A deep dive and comparison of Python drivers for Cassandra and Scylla Why and how we wrote a

A deep dive and comparison of Python drivers for Cassandra and Scylla Why and how we wrote a

A deep dive and comparison of Python drivers for Cassandra and Scylla Why and how we wrote a Python driver for Scylla EuroPython 2020 Bonjour ! Alexys Jacob Gentoo Linux developer - dev-db / mongodb / redis / scylla CTO at - sys-cluster

658 views • 42 slides
Environmental policy with intermittent sources of energy Stefan Ambec and Claude Crampes

Environmental policy with intermittent sources of energy Stefan Ambec and Claude Crampes

Environmental policy with intermittent sources of energy Stefan Ambec and Claude Crampes Toulouse School of Economics September 2015 Model Analysis Market power Motivation Intermittent sources of energy (wind, solar,...) Retail price

806 views • 34 slides
WINLAB Chenren Xu Joint work with Bernhard Firner, Yanyong Zhang Richard Howard, Jun Li,

WINLAB Chenren Xu Joint work with Bernhard Firner, Yanyong Zhang Richard Howard, Jun Li,

Improving RF-Based Device-Free Passive Localization Through Probabilistic Classification Methods Fall 2012 Research Review WINLAB Chenren Xu Joint work with Bernhard Firner, Yanyong Zhang Richard Howard, Jun Li, Xiaodong Lin Pervasive Radio

1.12k views • 66 slides
Shattering Fundamental Design Barriers of End-to-End Ultrahigh Data-Rate Transceivers: Direct

Shattering Fundamental Design Barriers of End-to-End Ultrahigh Data-Rate Transceivers: Direct

T T T T T T T T T T T T h h h h h h h h h h h h e e e e e e e e e e e e Shattering Fundamental Design Barriers of End-to-End Ultrahigh Data-Rate Transceivers: Direct Modulation in RF Domain Payam Heydari NCIC

776 views • 53 slides
GoBack A new view on phantom views Andreas Rieder Institut f ur Angewandte und Numerische

GoBack A new view on phantom views Andreas Rieder Institut f ur Angewandte und Numerische

GoBack A new view on phantom views Andreas Rieder Institut f ur Angewandte und Numerische Mathematik Universit at Karlsruhe Fakult at f ur Mathematik (jointly with Arne Schneck, Karlsruhe) c Andreas Rieder, Wien, AIP 09 1

476 views • 33 slides
Multimodal Framework Proposal Skip Cave Chief Scientist, Intervoice Inc. 1 Workshop Goals

Multimodal Framework Proposal Skip Cave Chief Scientist, Intervoice Inc. 1 Workshop Goals

Multimodal Framework Proposal Skip Cave Chief Scientist, Intervoice Inc. 1 Workshop Goals Identify & prioritize requirements for changes, extensions, and additions to the MMI architecture to better support Speech, GUI, Ink, and

188 views • 15 slides

More recommend