SlideShare una empresa de Scribd logo
1 de 45
Technology
An Agile
Approach to
Machine
Learning
Randy Shoup
VP Engineering
Background
@randyshoup
Technology
1. The Problem
What problem are
you trying to solve?
Agree on what you
are optimizing
Technology @randyshoup
• aka “Optimization Function” or “One
Metric That Matters”
• Discussing and agreeing on this metric
is itself valuable
• Only very few metrics, preferably one
Overall Evaluation
Criterion (OEC)
• E.g., Actions vs. click rate
• E.g., Long-term customer value vs.
short-term revenue
• “Pirate metrics” (AARRR): Acquisition,
Activation, Retention, Revenue,
Referral
Aligned to Business
Value
• Validated by data science, not solely
chosen by product / business
• Look for predictive leading indicators
• Avoid lagging indicators and vanity
metrics
Valid and
Measurable
Evaluating Success
Problem
“A problem
well-stated
is a problem
half-solved.”
-- Charles Kettering,
head of research at GM
Technology
Problem Difficulty
Problem
https://xkcd.com/1425/
Technology
2. The Data
Technology @randyshoup
• Many events, only predictive in
aggregate
• E.g., web search queries, ecommerce
clickstream, Netflix viewing metrics
Big but Shallow
• Few events, each of which is significant
• E.g., ecommerce purchases, WeWork
event attendance
Small but Deep
Characterizing Your Data
Data
Better data beats a
smarter algorithm
Technology @randyshoup
• Missing data, partial data
• Improperly or inconsistently formatted
Clean Data
• Consolidated into a single (logical)
location so it can be processed or
analyzed
• Joined together (“enriched”) with other
data sources
Aggregated Data
• Tagged by humans with one or more
labels
• Required to train supervised models
• Complicated and expensive at scale
Labeled Data
Better Data
Data
Technology @randyshoup
• More potentially useful attributes
• More data sources
• Longer retention
More Data
• Data pipeline to automate collection and
aggregation
• Move from large batch to mini-batch to
streaming data
Timely Data
Better Data
Data
“Data preparation accounts
for about 80% of the work of
data scientists.” – CrowdFlower survey,
2016
https://www.forbes.com/sites/gilpress/2016/03/23/data-preparation-most-time-consuming-least-enjoyable-data-science-task-survey-says/#2d58f4ab6f63
Technology
3. The Algorithms
Technology @randyshoup
• Encode expert knowledge
• Simple set of imperative if-then-else
statements
• Brittle and primitive
• Surprisingly effective
Rules and Heuristics
• Regression
• Decision trees / forests
• Collaborative filtering
• May be all you need
Simple Algorithms
• Iterative Optimization / Dynamic
Programming
• Neural nets
• Deep learning
• Only when absolutely required
Advanced Techniques
Algorithmic Evolution
Algorithms
Technology @randyshoup
• Many real-world problems are best
solved through a combination of several
algorithms
• E.g., Netflix Prize
Portfolio / Ensemble
Approaches
Algorithmic Evolution
Algorithms
Technology
Model
Execution
Online Model
Execution
Algorithms
Deploy Model
Collect Data
Train Model✅
Usage
@randyshoup
Technology
Offline Model
Building
Algorithms
Model
Execution
✅
Model
Building
Try New
Model
✅
@randyshoup
Technology @randyshoup
• Many common algorithms are highly
accurate, but difficult to interpret
• Model can make a decision, but ew
cannot “explain” its decision
• Particularly important in context of
system bias
• (+) Decision trees / forests, linear
regression
• (-) Neural nets, Deep Learning
Interpretability /
Explainability
• Enable data scientists to be self-
sufficient in experimenting, building,
training, and deploying
• End-to-end responsibility for models in
production
• Write models, deploy models, monitor
model performance
DevOps for
Data Science
• Platform-as-a-service for data scientists
• Programming model that matches the
workflow of a data scientist
• Abstract away infrastructure and other
details
Algorithm
Platform
Scaling Algorithm Development
Algorithms
Technology @randyshoup
• Data scientists spin up their own resources
• Both ad-hoc execution and repeatable pipelines
• Data science-friendly programming model exposes ETL and
Matrix transforms
• Abstracts away storage (S3), computation (Docker and ECS), and
the model building pipeline (Spark)
Algorithm Platform-as-a-Service
Algorithms
Technology
4. The Experiments
“It doesn’t matter how
beautiful your theory is.
It doesn’t matter how
smart you are.
If it doesn’t agree with
experiment, it’s wrong.”
-- Richard Feynman
Technology @randyshoup
• What metrics do you expect to move,
and why
• Understand your baseline
1. State Your
Hypothesis
• Sample size based on effect size
• Separate control and treatment groups,
test for bias
• Split traffic between control and
treatment
2. Design a Real A|B
Test
• Understand customer and system
behavior
• Understand why this experiment worked
or did not
3. Obsessively Log and
Measure
Designing and Running
Experimental Discipline
Technology @randyshoup
• Data trumps hope and intuition
• Develop insights for the next experiment
4. Listen to the
Data
• This is a journey, not a single step
5. Rinse and Repeat
Designing and Running
Experimental Discipline
Technology @randyshoup
Listen to the Data
Experimental Discipline
• 1/3 of ideas were positive and
statistically significant
• 1/3 of ideas were flat: no
statistically significant difference
• 1/3 of ideas were negative and
statistically significant
https://exp-platform.com/experiments-at-microsoft/
“Being wrong isn’t a bad
thing, like they teach
you in school. It is an
opportunity to learn
something.”
-- Richard Feynman
Technology @randyshoup
• Low-risk, push-button deployment
• Rapid release cadence
• Rapid rollback and recovery
Repeatable Deployment
Pipeline
• Faster to repair
• Easier to understand
• Simpler to diagnose
Smaller Units of Work
• Changes can be rolled out and rolled
back
• Learnings can be applied in the next
experiment
Enables
Experimentation
Continuous Delivery
Experimental Discipline
Technology @randyshoup
• Flag controls whether feature is “on” for
a particular set of users
• Independently discovered at eBay,
Yahoo, Google
• Decouple feature delivery from code
delivery
Enable / Disable feature
via configuration
• Develop / test / verify in production
• Rapid on or off for any reason
Makes Speed Safe
• Overall experiment controlled by feature
flag
• Control vs. treatment
Enables
Experimentation
Feature Flags
Experimental Discipline
● Ranking function for search results
○ Small number of hand-tuned factors  Thousands of factors
● Incremental Experimentation
○ Predictive models: query->view, view->purchase, etc.
○ Hundreds of parallel A | B tests
○ Full year of steady, incremental improvements
 2% increase in eBay revenue (~$120M / year)
@randyshoup
Machine-Learned Ranking
● Reduce user-experienced latency for search results
● Iterative Process
○ Implement a potential improvement
○ Release to the site in an A | B test
○ Monitor metrics –time to first byte, time to click, click rate, purchase rate
 2% increase in eBay revenue (~$120M / year)
@randyshoup
Site Speed
The most
dangerous
animal is the
“HiPPO”
Technology 33
Putting it All Together
Technology
Event Recommendations
WeWork Member Experience
Member Knowledge
Graph
Skills and
Interests
Event Feedback
Event Recommender
Predictive
Model
@randyshoup
Technology
Event Recipes
WeWork Member Experience
Event Recommender
Predictive
Model
@randyshoup
Technology
Get the predicted
opening occupancy
based on the
recommended 1-Click
price
Adjust the price to see how
occupancy will change
Occupancy Predictor
WeWork Revenue Optimization
@randyshoup
Technology
Revenue Simulation
WeWork Revenue Optimization
@randyshoup
Technology
Office Attributes Based Pricing
Corner office (premium)
Offices with high quality
views (premium)
Calculate and recommend
premium and discounts for
key office attributes
WeWork Revenue Optimization
@randyshoup
Technology
Example: Recommend alternative usage for unoccupied spaces
Fully optimize inventory usage by
leveraging demand and
profitability predictions
Inventory Management
WeWork Revenue Optimization
@randyshoup
Technology
Automatically lay out desk
configuration given space
constraints
Automated Layout
WeWork Applied Science
@randyshoup
Technology 41
Takeaways
Technology @randyshoup
• Identify and frame a clear business
problem
• … that matters to customers or the
business
• Define clear metric(s) for success
1. Drive from Business
Needs
• Single problem
• Solve problem end-to-end
• Show business results
2. Start Small
• Data collection and storage
• Data cleanliness and preparation
• Reliable, accurate, timely data pipeline
• Better data beats a better model (!)
3. Data Matters
Takeaways
An Agile Approach to Machine Learning
Technology @randyshoup
• Start with a Hypothesis
• Design an Experiment
• Separate Control and Experiment
group(s)
• Measure business metric for A vs. B
• Learn and Decide
4. A | B Testing
Discipline
• Simple model / No model
• Rules and Heuristics
• Gradually increase sophistication with
more data and more experience
5. Iteratively Refine
Model
• Find broader applicability across the
business
• Apply to more and more problems
• Move “upstream” in the development
process
6. Iteratively Expand
Applications
Takeaways
An Agile Approach to Machine Learning
Technology @randyshoup
• Make decisions with data instead of
guesswork and intuition
• Avoid HiPPO decisionmaking
• Can be threatening to designers,
product managers, decisionmakers
7. Data-Driven Culture
• Set of tools in our toolbox
• Sometimes valuable and useful
• Not a panacea
• Not a substitute for thinking 
8. Machine Learning is
not Magic
Takeaways
An Agile Approach to Machine Learning
Technology
New York
San Francisco
Tel Aviv
Shanghai
Singapore
Seattle
Palo Alto
Questions?
@randyshoup

Más contenido relacionado

La actualidad más candente

DOES15 - Randy Shoup - Ten (Hard-Won) Lessons of the DevOps Transition
DOES15 - Randy Shoup - Ten (Hard-Won) Lessons of the DevOps TransitionDOES15 - Randy Shoup - Ten (Hard-Won) Lessons of the DevOps Transition
DOES15 - Randy Shoup - Ten (Hard-Won) Lessons of the DevOps TransitionGene Kim
 
Why Enterprises Are Embracing the Cloud
Why Enterprises Are Embracing the CloudWhy Enterprises Are Embracing the Cloud
Why Enterprises Are Embracing the CloudRandy Shoup
 
Minimal Viable Architecture - Silicon Slopes 2020
Minimal Viable Architecture - Silicon Slopes 2020Minimal Viable Architecture - Silicon Slopes 2020
Minimal Viable Architecture - Silicon Slopes 2020Randy Shoup
 
Evolving Architecture and Organization - Lessons from Google and eBay
Evolving Architecture and Organization - Lessons from Google and eBayEvolving Architecture and Organization - Lessons from Google and eBay
Evolving Architecture and Organization - Lessons from Google and eBayRandy Shoup
 
A CTO's Guide to Scaling Organizations
A CTO's Guide to Scaling OrganizationsA CTO's Guide to Scaling Organizations
A CTO's Guide to Scaling OrganizationsRandy Shoup
 
Pragmatic Microservices
Pragmatic MicroservicesPragmatic Microservices
Pragmatic MicroservicesRandy Shoup
 
One Terrible Day at Google, and How It Made Us Better
One Terrible Day at Google, and How It Made Us BetterOne Terrible Day at Google, and How It Made Us Better
One Terrible Day at Google, and How It Made Us BetterRandy Shoup
 
DevOpsDays Silicon Valley 2014 - The Game of Operations
DevOpsDays Silicon Valley 2014 - The Game of OperationsDevOpsDays Silicon Valley 2014 - The Game of Operations
DevOpsDays Silicon Valley 2014 - The Game of OperationsRandy Shoup
 
Scaling Your Architecture with Services and Events
Scaling Your Architecture with Services and EventsScaling Your Architecture with Services and Events
Scaling Your Architecture with Services and EventsRandy Shoup
 
The Importance of Culture: Building and Sustaining Effective Engineering Org...
The Importance of Culture:  Building and Sustaining Effective Engineering Org...The Importance of Culture:  Building and Sustaining Effective Engineering Org...
The Importance of Culture: Building and Sustaining Effective Engineering Org...Randy Shoup
 
Learning from Learnings: Anatomy of Three Incidents
Learning from Learnings: Anatomy of Three IncidentsLearning from Learnings: Anatomy of Three Incidents
Learning from Learnings: Anatomy of Three IncidentsRandy Shoup
 
Managing Data at Scale - Microservices and Events
Managing Data at Scale - Microservices and EventsManaging Data at Scale - Microservices and Events
Managing Data at Scale - Microservices and EventsRandy Shoup
 
Anatomy of Three Incidents -- Commonalities and Lessons
Anatomy of Three Incidents -- Commonalities and LessonsAnatomy of Three Incidents -- Commonalities and Lessons
Anatomy of Three Incidents -- Commonalities and LessonsRandy Shoup
 
Flowcon2013 - Virtuous Cycles of Velocity: What I Learned About Going Fast at...
Flowcon2013 - Virtuous Cycles of Velocity: What I Learned About Going Fast at...Flowcon2013 - Virtuous Cycles of Velocity: What I Learned About Going Fast at...
Flowcon2013 - Virtuous Cycles of Velocity: What I Learned About Going Fast at...Randy Shoup
 
Teaching Machines to Fish -- How eBay Improves Itself
Teaching Machines to Fish -- How eBay Improves ItselfTeaching Machines to Fish -- How eBay Improves Itself
Teaching Machines to Fish -- How eBay Improves ItselfRandy Shoup
 
Tales from the Platform Trade
Tales from the Platform TradeTales from the Platform Trade
Tales from the Platform TradeWilliam Grosso
 
Serverless Toronto helps Startups
Serverless Toronto helps StartupsServerless Toronto helps Startups
Serverless Toronto helps StartupsDaniel Zivkovic
 
Velocity Conference NYC 2014 - Real World DevOps
Velocity Conference NYC 2014 - Real World DevOpsVelocity Conference NYC 2014 - Real World DevOps
Velocity Conference NYC 2014 - Real World DevOpsRodrigo Campos
 
2015 Mastering SAP Tech - Enterprise Mobility - Testing Lessons Learned
2015 Mastering SAP Tech - Enterprise Mobility - Testing Lessons Learned2015 Mastering SAP Tech - Enterprise Mobility - Testing Lessons Learned
2015 Mastering SAP Tech - Enterprise Mobility - Testing Lessons LearnedEneko Jon Bilbao
 
Supersize me: Making Drupal go large
Supersize me: Making Drupal go largeSupersize me: Making Drupal go large
Supersize me: Making Drupal go largeTom Phethean
 

La actualidad más candente (20)

DOES15 - Randy Shoup - Ten (Hard-Won) Lessons of the DevOps Transition
DOES15 - Randy Shoup - Ten (Hard-Won) Lessons of the DevOps TransitionDOES15 - Randy Shoup - Ten (Hard-Won) Lessons of the DevOps Transition
DOES15 - Randy Shoup - Ten (Hard-Won) Lessons of the DevOps Transition
 
Why Enterprises Are Embracing the Cloud
Why Enterprises Are Embracing the CloudWhy Enterprises Are Embracing the Cloud
Why Enterprises Are Embracing the Cloud
 
Minimal Viable Architecture - Silicon Slopes 2020
Minimal Viable Architecture - Silicon Slopes 2020Minimal Viable Architecture - Silicon Slopes 2020
Minimal Viable Architecture - Silicon Slopes 2020
 
Evolving Architecture and Organization - Lessons from Google and eBay
Evolving Architecture and Organization - Lessons from Google and eBayEvolving Architecture and Organization - Lessons from Google and eBay
Evolving Architecture and Organization - Lessons from Google and eBay
 
A CTO's Guide to Scaling Organizations
A CTO's Guide to Scaling OrganizationsA CTO's Guide to Scaling Organizations
A CTO's Guide to Scaling Organizations
 
Pragmatic Microservices
Pragmatic MicroservicesPragmatic Microservices
Pragmatic Microservices
 
One Terrible Day at Google, and How It Made Us Better
One Terrible Day at Google, and How It Made Us BetterOne Terrible Day at Google, and How It Made Us Better
One Terrible Day at Google, and How It Made Us Better
 
DevOpsDays Silicon Valley 2014 - The Game of Operations
DevOpsDays Silicon Valley 2014 - The Game of OperationsDevOpsDays Silicon Valley 2014 - The Game of Operations
DevOpsDays Silicon Valley 2014 - The Game of Operations
 
Scaling Your Architecture with Services and Events
Scaling Your Architecture with Services and EventsScaling Your Architecture with Services and Events
Scaling Your Architecture with Services and Events
 
The Importance of Culture: Building and Sustaining Effective Engineering Org...
The Importance of Culture:  Building and Sustaining Effective Engineering Org...The Importance of Culture:  Building and Sustaining Effective Engineering Org...
The Importance of Culture: Building and Sustaining Effective Engineering Org...
 
Learning from Learnings: Anatomy of Three Incidents
Learning from Learnings: Anatomy of Three IncidentsLearning from Learnings: Anatomy of Three Incidents
Learning from Learnings: Anatomy of Three Incidents
 
Managing Data at Scale - Microservices and Events
Managing Data at Scale - Microservices and EventsManaging Data at Scale - Microservices and Events
Managing Data at Scale - Microservices and Events
 
Anatomy of Three Incidents -- Commonalities and Lessons
Anatomy of Three Incidents -- Commonalities and LessonsAnatomy of Three Incidents -- Commonalities and Lessons
Anatomy of Three Incidents -- Commonalities and Lessons
 
Flowcon2013 - Virtuous Cycles of Velocity: What I Learned About Going Fast at...
Flowcon2013 - Virtuous Cycles of Velocity: What I Learned About Going Fast at...Flowcon2013 - Virtuous Cycles of Velocity: What I Learned About Going Fast at...
Flowcon2013 - Virtuous Cycles of Velocity: What I Learned About Going Fast at...
 
Teaching Machines to Fish -- How eBay Improves Itself
Teaching Machines to Fish -- How eBay Improves ItselfTeaching Machines to Fish -- How eBay Improves Itself
Teaching Machines to Fish -- How eBay Improves Itself
 
Tales from the Platform Trade
Tales from the Platform TradeTales from the Platform Trade
Tales from the Platform Trade
 
Serverless Toronto helps Startups
Serverless Toronto helps StartupsServerless Toronto helps Startups
Serverless Toronto helps Startups
 
Velocity Conference NYC 2014 - Real World DevOps
Velocity Conference NYC 2014 - Real World DevOpsVelocity Conference NYC 2014 - Real World DevOps
Velocity Conference NYC 2014 - Real World DevOps
 
2015 Mastering SAP Tech - Enterprise Mobility - Testing Lessons Learned
2015 Mastering SAP Tech - Enterprise Mobility - Testing Lessons Learned2015 Mastering SAP Tech - Enterprise Mobility - Testing Lessons Learned
2015 Mastering SAP Tech - Enterprise Mobility - Testing Lessons Learned
 
Supersize me: Making Drupal go large
Supersize me: Making Drupal go largeSupersize me: Making Drupal go large
Supersize me: Making Drupal go large
 

Similar a An Agile Approach to Machine Learning

Productionising Machine Learning Models
Productionising Machine Learning ModelsProductionising Machine Learning Models
Productionising Machine Learning ModelsTash Bickley
 
Mistakes we make_and_howto_avoid_them_v0.12
Mistakes we make_and_howto_avoid_them_v0.12Mistakes we make_and_howto_avoid_them_v0.12
Mistakes we make_and_howto_avoid_them_v0.12Trevor Warren
 
Can we induce change with what we measure?
Can we induce change with what we measure?Can we induce change with what we measure?
Can we induce change with what we measure?Michaela Greiler
 
Big Data at a Gaming Company: Spil Games
Big Data at a Gaming Company: Spil GamesBig Data at a Gaming Company: Spil Games
Big Data at a Gaming Company: Spil GamesRob Winters
 
Doing Analytics Right - Building the Analytics Environment
Doing Analytics Right - Building the Analytics EnvironmentDoing Analytics Right - Building the Analytics Environment
Doing Analytics Right - Building the Analytics EnvironmentTasktop
 
The Automation Firehose: Be Strategic & Tactical With Your Mobile & Web Testing
The Automation Firehose: Be Strategic & Tactical With Your Mobile & Web TestingThe Automation Firehose: Be Strategic & Tactical With Your Mobile & Web Testing
The Automation Firehose: Be Strategic & Tactical With Your Mobile & Web TestingPerfecto by Perforce
 
Lean Startup: Reduce 40% go-to-market time & cost on your next product launch
Lean Startup: Reduce 40% go-to-market time & cost on your next product launchLean Startup: Reduce 40% go-to-market time & cost on your next product launch
Lean Startup: Reduce 40% go-to-market time & cost on your next product launchPeople10 Technosoft Private Limited
 
Ericriesleanstartuppresentationforweb2
Ericriesleanstartuppresentationforweb2Ericriesleanstartuppresentationforweb2
Ericriesleanstartuppresentationforweb2Edmund FOng
 
How to Use Artificial Intelligence by Microsoft Product Manager
 How to Use Artificial Intelligence by Microsoft Product Manager How to Use Artificial Intelligence by Microsoft Product Manager
How to Use Artificial Intelligence by Microsoft Product ManagerProduct School
 
PAC 2019 virtual Alexander Podelko
PAC 2019 virtual Alexander Podelko PAC 2019 virtual Alexander Podelko
PAC 2019 virtual Alexander Podelko Neotys
 
R+Hadoop - Ask Bigger (and New) Questions and Get Better, Faster Answers
R+Hadoop - Ask Bigger (and New) Questions and Get Better, Faster AnswersR+Hadoop - Ask Bigger (and New) Questions and Get Better, Faster Answers
R+Hadoop - Ask Bigger (and New) Questions and Get Better, Faster AnswersRevolution Analytics
 
FlorenceAI: Reinventing Data Science at Humana
FlorenceAI: Reinventing Data Science at HumanaFlorenceAI: Reinventing Data Science at Humana
FlorenceAI: Reinventing Data Science at HumanaDatabricks
 
Building an Open Source AppSec Pipeline
Building an Open Source AppSec PipelineBuilding an Open Source AppSec Pipeline
Building an Open Source AppSec PipelineMatt Tesauro
 
Lifecycle of a Data Science Project
Lifecycle of a Data Science ProjectLifecycle of a Data Science Project
Lifecycle of a Data Science ProjectDigital Vidya
 
New Model Testing: A New Test Process and Tool
New Model Testing:  A New Test Process and ToolNew Model Testing:  A New Test Process and Tool
New Model Testing: A New Test Process and ToolTEST Huddle
 
Eureka Data Science Analytic Process
Eureka Data Science Analytic ProcessEureka Data Science Analytic Process
Eureka Data Science Analytic ProcessAllen Nugent
 
Alexander Podelko - Context-Driven Performance Testing
Alexander Podelko - Context-Driven Performance TestingAlexander Podelko - Context-Driven Performance Testing
Alexander Podelko - Context-Driven Performance TestingNeotys_Partner
 
How Celtra Optimizes its Advertising Platform with Databricks
How Celtra Optimizes its Advertising Platformwith DatabricksHow Celtra Optimizes its Advertising Platformwith Databricks
How Celtra Optimizes its Advertising Platform with DatabricksGrega Kespret
 
Building and Scaling High Performing Technology Organizations by Jez Humble a...
Building and Scaling High Performing Technology Organizations by Jez Humble a...Building and Scaling High Performing Technology Organizations by Jez Humble a...
Building and Scaling High Performing Technology Organizations by Jez Humble a...Agile India
 

Similar a An Agile Approach to Machine Learning (20)

Productionising Machine Learning Models
Productionising Machine Learning ModelsProductionising Machine Learning Models
Productionising Machine Learning Models
 
Mistakes we make_and_howto_avoid_them_v0.12
Mistakes we make_and_howto_avoid_them_v0.12Mistakes we make_and_howto_avoid_them_v0.12
Mistakes we make_and_howto_avoid_them_v0.12
 
Can we induce change with what we measure?
Can we induce change with what we measure?Can we induce change with what we measure?
Can we induce change with what we measure?
 
Big Data at a Gaming Company: Spil Games
Big Data at a Gaming Company: Spil GamesBig Data at a Gaming Company: Spil Games
Big Data at a Gaming Company: Spil Games
 
Doing Analytics Right - Building the Analytics Environment
Doing Analytics Right - Building the Analytics EnvironmentDoing Analytics Right - Building the Analytics Environment
Doing Analytics Right - Building the Analytics Environment
 
The Automation Firehose: Be Strategic & Tactical With Your Mobile & Web Testing
The Automation Firehose: Be Strategic & Tactical With Your Mobile & Web TestingThe Automation Firehose: Be Strategic & Tactical With Your Mobile & Web Testing
The Automation Firehose: Be Strategic & Tactical With Your Mobile & Web Testing
 
Lean Startup: Reduce 40% go-to-market time & cost on your next product launch
Lean Startup: Reduce 40% go-to-market time & cost on your next product launchLean Startup: Reduce 40% go-to-market time & cost on your next product launch
Lean Startup: Reduce 40% go-to-market time & cost on your next product launch
 
Ericriesleanstartuppresentationforweb2
Ericriesleanstartuppresentationforweb2Ericriesleanstartuppresentationforweb2
Ericriesleanstartuppresentationforweb2
 
How to Use Artificial Intelligence by Microsoft Product Manager
 How to Use Artificial Intelligence by Microsoft Product Manager How to Use Artificial Intelligence by Microsoft Product Manager
How to Use Artificial Intelligence by Microsoft Product Manager
 
PAC 2019 virtual Alexander Podelko
PAC 2019 virtual Alexander Podelko PAC 2019 virtual Alexander Podelko
PAC 2019 virtual Alexander Podelko
 
R+Hadoop - Ask Bigger (and New) Questions and Get Better, Faster Answers
R+Hadoop - Ask Bigger (and New) Questions and Get Better, Faster AnswersR+Hadoop - Ask Bigger (and New) Questions and Get Better, Faster Answers
R+Hadoop - Ask Bigger (and New) Questions and Get Better, Faster Answers
 
FlorenceAI: Reinventing Data Science at Humana
FlorenceAI: Reinventing Data Science at HumanaFlorenceAI: Reinventing Data Science at Humana
FlorenceAI: Reinventing Data Science at Humana
 
Building an Open Source AppSec Pipeline
Building an Open Source AppSec PipelineBuilding an Open Source AppSec Pipeline
Building an Open Source AppSec Pipeline
 
Training - What is Performance ?
Training  - What is Performance ?Training  - What is Performance ?
Training - What is Performance ?
 
Lifecycle of a Data Science Project
Lifecycle of a Data Science ProjectLifecycle of a Data Science Project
Lifecycle of a Data Science Project
 
New Model Testing: A New Test Process and Tool
New Model Testing:  A New Test Process and ToolNew Model Testing:  A New Test Process and Tool
New Model Testing: A New Test Process and Tool
 
Eureka Data Science Analytic Process
Eureka Data Science Analytic ProcessEureka Data Science Analytic Process
Eureka Data Science Analytic Process
 
Alexander Podelko - Context-Driven Performance Testing
Alexander Podelko - Context-Driven Performance TestingAlexander Podelko - Context-Driven Performance Testing
Alexander Podelko - Context-Driven Performance Testing
 
How Celtra Optimizes its Advertising Platform with Databricks
How Celtra Optimizes its Advertising Platformwith DatabricksHow Celtra Optimizes its Advertising Platformwith Databricks
How Celtra Optimizes its Advertising Platform with Databricks
 
Building and Scaling High Performing Technology Organizations by Jez Humble a...
Building and Scaling High Performing Technology Organizations by Jez Humble a...Building and Scaling High Performing Technology Organizations by Jez Humble a...
Building and Scaling High Performing Technology Organizations by Jez Humble a...
 

Más de Randy Shoup

Large Scale Architecture -- The Unreasonable Effectiveness of Simplicity
Large Scale Architecture -- The Unreasonable Effectiveness of SimplicityLarge Scale Architecture -- The Unreasonable Effectiveness of Simplicity
Large Scale Architecture -- The Unreasonable Effectiveness of SimplicityRandy Shoup
 
Breaking Codes, Designing Jets, and Building Teams
Breaking Codes, Designing Jets, and Building TeamsBreaking Codes, Designing Jets, and Building Teams
Breaking Codes, Designing Jets, and Building TeamsRandy Shoup
 
Monoliths, Migrations, and Microservices
Monoliths, Migrations, and MicroservicesMonoliths, Migrations, and Microservices
Monoliths, Migrations, and MicroservicesRandy Shoup
 
Ten Lessons of the DevOps Transition
Ten Lessons of the DevOps TransitionTen Lessons of the DevOps Transition
Ten Lessons of the DevOps TransitionRandy Shoup
 
Managing Data in Microservices
Managing Data in MicroservicesManaging Data in Microservices
Managing Data in MicroservicesRandy Shoup
 
Effective Microservices In a Data-centric World
Effective Microservices In a Data-centric WorldEffective Microservices In a Data-centric World
Effective Microservices In a Data-centric WorldRandy Shoup
 
From the Monolith to Microservices - CraftConf 2015
From the Monolith to Microservices - CraftConf 2015From the Monolith to Microservices - CraftConf 2015
From the Monolith to Microservices - CraftConf 2015Randy Shoup
 
Concurrency at Scale: Evolution to Micro-Services
Concurrency at Scale:  Evolution to Micro-ServicesConcurrency at Scale:  Evolution to Micro-Services
Concurrency at Scale: Evolution to Micro-ServicesRandy Shoup
 
QCon New York 2014 - Scalable, Reliable Analytics Infrastructure at KIXEYE
QCon New York 2014 - Scalable, Reliable Analytics Infrastructure at KIXEYEQCon New York 2014 - Scalable, Reliable Analytics Infrastructure at KIXEYE
QCon New York 2014 - Scalable, Reliable Analytics Infrastructure at KIXEYERandy Shoup
 
QCon Tokyo 2014 - Virtuous Cycles of Velocity: What I Learned About Going Fas...
QCon Tokyo 2014 - Virtuous Cycles of Velocity: What I Learned About Going Fas...QCon Tokyo 2014 - Virtuous Cycles of Velocity: What I Learned About Going Fas...
QCon Tokyo 2014 - Virtuous Cycles of Velocity: What I Learned About Going Fas...Randy Shoup
 

Más de Randy Shoup (10)

Large Scale Architecture -- The Unreasonable Effectiveness of Simplicity
Large Scale Architecture -- The Unreasonable Effectiveness of SimplicityLarge Scale Architecture -- The Unreasonable Effectiveness of Simplicity
Large Scale Architecture -- The Unreasonable Effectiveness of Simplicity
 
Breaking Codes, Designing Jets, and Building Teams
Breaking Codes, Designing Jets, and Building TeamsBreaking Codes, Designing Jets, and Building Teams
Breaking Codes, Designing Jets, and Building Teams
 
Monoliths, Migrations, and Microservices
Monoliths, Migrations, and MicroservicesMonoliths, Migrations, and Microservices
Monoliths, Migrations, and Microservices
 
Ten Lessons of the DevOps Transition
Ten Lessons of the DevOps TransitionTen Lessons of the DevOps Transition
Ten Lessons of the DevOps Transition
 
Managing Data in Microservices
Managing Data in MicroservicesManaging Data in Microservices
Managing Data in Microservices
 
Effective Microservices In a Data-centric World
Effective Microservices In a Data-centric WorldEffective Microservices In a Data-centric World
Effective Microservices In a Data-centric World
 
From the Monolith to Microservices - CraftConf 2015
From the Monolith to Microservices - CraftConf 2015From the Monolith to Microservices - CraftConf 2015
From the Monolith to Microservices - CraftConf 2015
 
Concurrency at Scale: Evolution to Micro-Services
Concurrency at Scale:  Evolution to Micro-ServicesConcurrency at Scale:  Evolution to Micro-Services
Concurrency at Scale: Evolution to Micro-Services
 
QCon New York 2014 - Scalable, Reliable Analytics Infrastructure at KIXEYE
QCon New York 2014 - Scalable, Reliable Analytics Infrastructure at KIXEYEQCon New York 2014 - Scalable, Reliable Analytics Infrastructure at KIXEYE
QCon New York 2014 - Scalable, Reliable Analytics Infrastructure at KIXEYE
 
QCon Tokyo 2014 - Virtuous Cycles of Velocity: What I Learned About Going Fas...
QCon Tokyo 2014 - Virtuous Cycles of Velocity: What I Learned About Going Fas...QCon Tokyo 2014 - Virtuous Cycles of Velocity: What I Learned About Going Fas...
QCon Tokyo 2014 - Virtuous Cycles of Velocity: What I Learned About Going Fas...
 

Último

5 Signs You Need a Fashion PLM Software.pdf
5 Signs You Need a Fashion PLM Software.pdf5 Signs You Need a Fashion PLM Software.pdf
5 Signs You Need a Fashion PLM Software.pdfWave PLM
 
Reassessing the Bedrock of Clinical Function Models: An Examination of Large ...
Reassessing the Bedrock of Clinical Function Models: An Examination of Large ...Reassessing the Bedrock of Clinical Function Models: An Examination of Large ...
Reassessing the Bedrock of Clinical Function Models: An Examination of Large ...harshavardhanraghave
 
Optimizing AI for immediate response in Smart CCTV
Optimizing AI for immediate response in Smart CCTVOptimizing AI for immediate response in Smart CCTV
Optimizing AI for immediate response in Smart CCTVshikhaohhpro
 
SyndBuddy AI 2k Review 2024: Revolutionizing Content Syndication with AI
SyndBuddy AI 2k Review 2024: Revolutionizing Content Syndication with AISyndBuddy AI 2k Review 2024: Revolutionizing Content Syndication with AI
SyndBuddy AI 2k Review 2024: Revolutionizing Content Syndication with AIABDERRAOUF MEHENNI
 
Short Story: Unveiling the Reasoning Abilities of Large Language Models by Ke...
Short Story: Unveiling the Reasoning Abilities of Large Language Models by Ke...Short Story: Unveiling the Reasoning Abilities of Large Language Models by Ke...
Short Story: Unveiling the Reasoning Abilities of Large Language Models by Ke...kellynguyen01
 
How To Use Server-Side Rendering with Nuxt.js
How To Use Server-Side Rendering with Nuxt.jsHow To Use Server-Side Rendering with Nuxt.js
How To Use Server-Side Rendering with Nuxt.jsAndolasoft Inc
 
Tech Tuesday-Harness the Power of Effective Resource Planning with OnePlan’s ...
Tech Tuesday-Harness the Power of Effective Resource Planning with OnePlan’s ...Tech Tuesday-Harness the Power of Effective Resource Planning with OnePlan’s ...
Tech Tuesday-Harness the Power of Effective Resource Planning with OnePlan’s ...OnePlan Solutions
 
TECUNIQUE: Success Stories: IT Service provider
TECUNIQUE: Success Stories: IT Service providerTECUNIQUE: Success Stories: IT Service provider
TECUNIQUE: Success Stories: IT Service providermohitmore19
 
How To Troubleshoot Collaboration Apps for the Modern Connected Worker
How To Troubleshoot Collaboration Apps for the Modern Connected WorkerHow To Troubleshoot Collaboration Apps for the Modern Connected Worker
How To Troubleshoot Collaboration Apps for the Modern Connected WorkerThousandEyes
 
call girls in Vaishali (Ghaziabad) 🔝 >༒8448380779 🔝 genuine Escort Service 🔝✔️✔️
call girls in Vaishali (Ghaziabad) 🔝 >༒8448380779 🔝 genuine Escort Service 🔝✔️✔️call girls in Vaishali (Ghaziabad) 🔝 >༒8448380779 🔝 genuine Escort Service 🔝✔️✔️
call girls in Vaishali (Ghaziabad) 🔝 >༒8448380779 🔝 genuine Escort Service 🔝✔️✔️Delhi Call girls
 
Diamond Application Development Crafting Solutions with Precision
Diamond Application Development Crafting Solutions with PrecisionDiamond Application Development Crafting Solutions with Precision
Diamond Application Development Crafting Solutions with PrecisionSolGuruz
 
CALL ON ➥8923113531 🔝Call Girls Kakori Lucknow best sexual service Online ☂️
CALL ON ➥8923113531 🔝Call Girls Kakori Lucknow best sexual service Online  ☂️CALL ON ➥8923113531 🔝Call Girls Kakori Lucknow best sexual service Online  ☂️
CALL ON ➥8923113531 🔝Call Girls Kakori Lucknow best sexual service Online ☂️anilsa9823
 
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...ICS
 
Software Quality Assurance Interview Questions
Software Quality Assurance Interview QuestionsSoftware Quality Assurance Interview Questions
Software Quality Assurance Interview QuestionsArshad QA
 
The Ultimate Test Automation Guide_ Best Practices and Tips.pdf
The Ultimate Test Automation Guide_ Best Practices and Tips.pdfThe Ultimate Test Automation Guide_ Best Practices and Tips.pdf
The Ultimate Test Automation Guide_ Best Practices and Tips.pdfkalichargn70th171
 
W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...
W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...
W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...panagenda
 
Right Money Management App For Your Financial Goals
Right Money Management App For Your Financial GoalsRight Money Management App For Your Financial Goals
Right Money Management App For Your Financial GoalsJhone kinadey
 
HR Software Buyers Guide in 2024 - HRSoftware.com
HR Software Buyers Guide in 2024 - HRSoftware.comHR Software Buyers Guide in 2024 - HRSoftware.com
HR Software Buyers Guide in 2024 - HRSoftware.comFatema Valibhai
 

Último (20)

5 Signs You Need a Fashion PLM Software.pdf
5 Signs You Need a Fashion PLM Software.pdf5 Signs You Need a Fashion PLM Software.pdf
5 Signs You Need a Fashion PLM Software.pdf
 
Reassessing the Bedrock of Clinical Function Models: An Examination of Large ...
Reassessing the Bedrock of Clinical Function Models: An Examination of Large ...Reassessing the Bedrock of Clinical Function Models: An Examination of Large ...
Reassessing the Bedrock of Clinical Function Models: An Examination of Large ...
 
Optimizing AI for immediate response in Smart CCTV
Optimizing AI for immediate response in Smart CCTVOptimizing AI for immediate response in Smart CCTV
Optimizing AI for immediate response in Smart CCTV
 
SyndBuddy AI 2k Review 2024: Revolutionizing Content Syndication with AI
SyndBuddy AI 2k Review 2024: Revolutionizing Content Syndication with AISyndBuddy AI 2k Review 2024: Revolutionizing Content Syndication with AI
SyndBuddy AI 2k Review 2024: Revolutionizing Content Syndication with AI
 
Short Story: Unveiling the Reasoning Abilities of Large Language Models by Ke...
Short Story: Unveiling the Reasoning Abilities of Large Language Models by Ke...Short Story: Unveiling the Reasoning Abilities of Large Language Models by Ke...
Short Story: Unveiling the Reasoning Abilities of Large Language Models by Ke...
 
How To Use Server-Side Rendering with Nuxt.js
How To Use Server-Side Rendering with Nuxt.jsHow To Use Server-Side Rendering with Nuxt.js
How To Use Server-Side Rendering with Nuxt.js
 
CHEAP Call Girls in Pushp Vihar (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICE
CHEAP Call Girls in Pushp Vihar (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICECHEAP Call Girls in Pushp Vihar (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICE
CHEAP Call Girls in Pushp Vihar (-DELHI )🔝 9953056974🔝(=)/CALL GIRLS SERVICE
 
Tech Tuesday-Harness the Power of Effective Resource Planning with OnePlan’s ...
Tech Tuesday-Harness the Power of Effective Resource Planning with OnePlan’s ...Tech Tuesday-Harness the Power of Effective Resource Planning with OnePlan’s ...
Tech Tuesday-Harness the Power of Effective Resource Planning with OnePlan’s ...
 
TECUNIQUE: Success Stories: IT Service provider
TECUNIQUE: Success Stories: IT Service providerTECUNIQUE: Success Stories: IT Service provider
TECUNIQUE: Success Stories: IT Service provider
 
How To Troubleshoot Collaboration Apps for the Modern Connected Worker
How To Troubleshoot Collaboration Apps for the Modern Connected WorkerHow To Troubleshoot Collaboration Apps for the Modern Connected Worker
How To Troubleshoot Collaboration Apps for the Modern Connected Worker
 
call girls in Vaishali (Ghaziabad) 🔝 >༒8448380779 🔝 genuine Escort Service 🔝✔️✔️
call girls in Vaishali (Ghaziabad) 🔝 >༒8448380779 🔝 genuine Escort Service 🔝✔️✔️call girls in Vaishali (Ghaziabad) 🔝 >༒8448380779 🔝 genuine Escort Service 🔝✔️✔️
call girls in Vaishali (Ghaziabad) 🔝 >༒8448380779 🔝 genuine Escort Service 🔝✔️✔️
 
Diamond Application Development Crafting Solutions with Precision
Diamond Application Development Crafting Solutions with PrecisionDiamond Application Development Crafting Solutions with Precision
Diamond Application Development Crafting Solutions with Precision
 
CALL ON ➥8923113531 🔝Call Girls Kakori Lucknow best sexual service Online ☂️
CALL ON ➥8923113531 🔝Call Girls Kakori Lucknow best sexual service Online  ☂️CALL ON ➥8923113531 🔝Call Girls Kakori Lucknow best sexual service Online  ☂️
CALL ON ➥8923113531 🔝Call Girls Kakori Lucknow best sexual service Online ☂️
 
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...
The Real-World Challenges of Medical Device Cybersecurity- Mitigating Vulnera...
 
Software Quality Assurance Interview Questions
Software Quality Assurance Interview QuestionsSoftware Quality Assurance Interview Questions
Software Quality Assurance Interview Questions
 
The Ultimate Test Automation Guide_ Best Practices and Tips.pdf
The Ultimate Test Automation Guide_ Best Practices and Tips.pdfThe Ultimate Test Automation Guide_ Best Practices and Tips.pdf
The Ultimate Test Automation Guide_ Best Practices and Tips.pdf
 
Microsoft AI Transformation Partner Playbook.pdf
Microsoft AI Transformation Partner Playbook.pdfMicrosoft AI Transformation Partner Playbook.pdf
Microsoft AI Transformation Partner Playbook.pdf
 
W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...
W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...
W01_panagenda_Navigating-the-Future-with-The-Hitchhikers-Guide-to-Notes-and-D...
 
Right Money Management App For Your Financial Goals
Right Money Management App For Your Financial GoalsRight Money Management App For Your Financial Goals
Right Money Management App For Your Financial Goals
 
HR Software Buyers Guide in 2024 - HRSoftware.com
HR Software Buyers Guide in 2024 - HRSoftware.comHR Software Buyers Guide in 2024 - HRSoftware.com
HR Software Buyers Guide in 2024 - HRSoftware.com
 

An Agile Approach to Machine Learning

  • 4. What problem are you trying to solve?
  • 5. Agree on what you are optimizing
  • 6. Technology @randyshoup • aka “Optimization Function” or “One Metric That Matters” • Discussing and agreeing on this metric is itself valuable • Only very few metrics, preferably one Overall Evaluation Criterion (OEC) • E.g., Actions vs. click rate • E.g., Long-term customer value vs. short-term revenue • “Pirate metrics” (AARRR): Acquisition, Activation, Retention, Revenue, Referral Aligned to Business Value • Validated by data science, not solely chosen by product / business • Look for predictive leading indicators • Avoid lagging indicators and vanity metrics Valid and Measurable Evaluating Success Problem
  • 7. “A problem well-stated is a problem half-solved.” -- Charles Kettering, head of research at GM
  • 10. Technology @randyshoup • Many events, only predictive in aggregate • E.g., web search queries, ecommerce clickstream, Netflix viewing metrics Big but Shallow • Few events, each of which is significant • E.g., ecommerce purchases, WeWork event attendance Small but Deep Characterizing Your Data Data
  • 11. Better data beats a smarter algorithm
  • 12. Technology @randyshoup • Missing data, partial data • Improperly or inconsistently formatted Clean Data • Consolidated into a single (logical) location so it can be processed or analyzed • Joined together (“enriched”) with other data sources Aggregated Data • Tagged by humans with one or more labels • Required to train supervised models • Complicated and expensive at scale Labeled Data Better Data Data
  • 13. Technology @randyshoup • More potentially useful attributes • More data sources • Longer retention More Data • Data pipeline to automate collection and aggregation • Move from large batch to mini-batch to streaming data Timely Data Better Data Data
  • 14. “Data preparation accounts for about 80% of the work of data scientists.” – CrowdFlower survey, 2016 https://www.forbes.com/sites/gilpress/2016/03/23/data-preparation-most-time-consuming-least-enjoyable-data-science-task-survey-says/#2d58f4ab6f63
  • 16. Technology @randyshoup • Encode expert knowledge • Simple set of imperative if-then-else statements • Brittle and primitive • Surprisingly effective Rules and Heuristics • Regression • Decision trees / forests • Collaborative filtering • May be all you need Simple Algorithms • Iterative Optimization / Dynamic Programming • Neural nets • Deep learning • Only when absolutely required Advanced Techniques Algorithmic Evolution Algorithms
  • 17. Technology @randyshoup • Many real-world problems are best solved through a combination of several algorithms • E.g., Netflix Prize Portfolio / Ensemble Approaches Algorithmic Evolution Algorithms
  • 20. Technology @randyshoup • Many common algorithms are highly accurate, but difficult to interpret • Model can make a decision, but ew cannot “explain” its decision • Particularly important in context of system bias • (+) Decision trees / forests, linear regression • (-) Neural nets, Deep Learning Interpretability / Explainability • Enable data scientists to be self- sufficient in experimenting, building, training, and deploying • End-to-end responsibility for models in production • Write models, deploy models, monitor model performance DevOps for Data Science • Platform-as-a-service for data scientists • Programming model that matches the workflow of a data scientist • Abstract away infrastructure and other details Algorithm Platform Scaling Algorithm Development Algorithms
  • 21. Technology @randyshoup • Data scientists spin up their own resources • Both ad-hoc execution and repeatable pipelines • Data science-friendly programming model exposes ETL and Matrix transforms • Abstracts away storage (S3), computation (Docker and ECS), and the model building pipeline (Spark) Algorithm Platform-as-a-Service Algorithms
  • 23. “It doesn’t matter how beautiful your theory is. It doesn’t matter how smart you are. If it doesn’t agree with experiment, it’s wrong.” -- Richard Feynman
  • 24. Technology @randyshoup • What metrics do you expect to move, and why • Understand your baseline 1. State Your Hypothesis • Sample size based on effect size • Separate control and treatment groups, test for bias • Split traffic between control and treatment 2. Design a Real A|B Test • Understand customer and system behavior • Understand why this experiment worked or did not 3. Obsessively Log and Measure Designing and Running Experimental Discipline
  • 25. Technology @randyshoup • Data trumps hope and intuition • Develop insights for the next experiment 4. Listen to the Data • This is a journey, not a single step 5. Rinse and Repeat Designing and Running Experimental Discipline
  • 26. Technology @randyshoup Listen to the Data Experimental Discipline • 1/3 of ideas were positive and statistically significant • 1/3 of ideas were flat: no statistically significant difference • 1/3 of ideas were negative and statistically significant https://exp-platform.com/experiments-at-microsoft/
  • 27. “Being wrong isn’t a bad thing, like they teach you in school. It is an opportunity to learn something.” -- Richard Feynman
  • 28. Technology @randyshoup • Low-risk, push-button deployment • Rapid release cadence • Rapid rollback and recovery Repeatable Deployment Pipeline • Faster to repair • Easier to understand • Simpler to diagnose Smaller Units of Work • Changes can be rolled out and rolled back • Learnings can be applied in the next experiment Enables Experimentation Continuous Delivery Experimental Discipline
  • 29. Technology @randyshoup • Flag controls whether feature is “on” for a particular set of users • Independently discovered at eBay, Yahoo, Google • Decouple feature delivery from code delivery Enable / Disable feature via configuration • Develop / test / verify in production • Rapid on or off for any reason Makes Speed Safe • Overall experiment controlled by feature flag • Control vs. treatment Enables Experimentation Feature Flags Experimental Discipline
  • 30. ● Ranking function for search results ○ Small number of hand-tuned factors  Thousands of factors ● Incremental Experimentation ○ Predictive models: query->view, view->purchase, etc. ○ Hundreds of parallel A | B tests ○ Full year of steady, incremental improvements  2% increase in eBay revenue (~$120M / year) @randyshoup Machine-Learned Ranking
  • 31. ● Reduce user-experienced latency for search results ● Iterative Process ○ Implement a potential improvement ○ Release to the site in an A | B test ○ Monitor metrics –time to first byte, time to click, click rate, purchase rate  2% increase in eBay revenue (~$120M / year) @randyshoup Site Speed
  • 32. The most dangerous animal is the “HiPPO”
  • 33. Technology 33 Putting it All Together
  • 34. Technology Event Recommendations WeWork Member Experience Member Knowledge Graph Skills and Interests Event Feedback Event Recommender Predictive Model @randyshoup
  • 35. Technology Event Recipes WeWork Member Experience Event Recommender Predictive Model @randyshoup
  • 36. Technology Get the predicted opening occupancy based on the recommended 1-Click price Adjust the price to see how occupancy will change Occupancy Predictor WeWork Revenue Optimization @randyshoup
  • 38. Technology Office Attributes Based Pricing Corner office (premium) Offices with high quality views (premium) Calculate and recommend premium and discounts for key office attributes WeWork Revenue Optimization @randyshoup
  • 39. Technology Example: Recommend alternative usage for unoccupied spaces Fully optimize inventory usage by leveraging demand and profitability predictions Inventory Management WeWork Revenue Optimization @randyshoup
  • 40. Technology Automatically lay out desk configuration given space constraints Automated Layout WeWork Applied Science @randyshoup
  • 42. Technology @randyshoup • Identify and frame a clear business problem • … that matters to customers or the business • Define clear metric(s) for success 1. Drive from Business Needs • Single problem • Solve problem end-to-end • Show business results 2. Start Small • Data collection and storage • Data cleanliness and preparation • Reliable, accurate, timely data pipeline • Better data beats a better model (!) 3. Data Matters Takeaways An Agile Approach to Machine Learning
  • 43. Technology @randyshoup • Start with a Hypothesis • Design an Experiment • Separate Control and Experiment group(s) • Measure business metric for A vs. B • Learn and Decide 4. A | B Testing Discipline • Simple model / No model • Rules and Heuristics • Gradually increase sophistication with more data and more experience 5. Iteratively Refine Model • Find broader applicability across the business • Apply to more and more problems • Move “upstream” in the development process 6. Iteratively Expand Applications Takeaways An Agile Approach to Machine Learning
  • 44. Technology @randyshoup • Make decisions with data instead of guesswork and intuition • Avoid HiPPO decisionmaking • Can be threatening to designers, product managers, decisionmakers 7. Data-Driven Culture • Set of tools in our toolbox • Sometimes valuable and useful • Not a panacea • Not a substitute for thinking  8. Machine Learning is not Magic Takeaways An Agile Approach to Machine Learning
  • 45. Technology New York San Francisco Tel Aviv Shanghai Singapore Seattle Palo Alto Questions? @randyshoup