Future of AI-powered automation in business

transcript

@louisdorard #APIdays - December 9, 2015

AI is everywhere

@louisdorard

Lars Trieloff

@trieloff

(see source)

Amazon for David Jones (@d_jones, see source)

ChurnSpotter.io

How does it work?

Data + Machine Learning

Bedrooms Bathrooms Surface (foot²) Year built Type Price ($)

3 1 860 1950 house 565,0003 1 1012 1951 house2 1.5 968 1976 townhouse 447,0004 1315 1950 house 648,0003 2 1599 1964 house3 2 987 1951 townhouse 790,0001 1 530 2007 condo 122,0004 2 1574 1964 house 835,0004 2001 house 855,0003 2.5 1472 2005 house4 3.5 1714 2005 townhouse2 2 1113 1999 condo1 769 1999 condo 315,000

Bedrooms Bathrooms Surface (foot²) Year built Type Price ($)

3 1 860 1950 house 565,0003 1 1012 1951 house2 1.5 968 1976 townhouse 447,0004 1315 1950 house 648,0003 2 1599 1964 house3 2 987 1951 townhouse 790,0001 1 530 2007 condo 122,0004 2 1574 1964 house 835,0004 2001 house 855,0003 2.5 1472 2005 house4 3.5 1714 2005 townhouse2 2 1113 1999 condo1 769 1999 condo 315,000

ML is a set of AI techniques where “intelligence” is built by referring to

examples

“Weak AI” vs. “Strong AI”

Ever yday use cases

• Real-estate

• Spam

• Priority inbox

• Crowd prediction

property price

email spam indicator

email importance indicator

location & context #people

Zillow

Tranquilien

Business use cases

• Reduce churn

• Cross-sell

• Optimize pricing

• Predict demand

customer churn indicator

customer & product purchase indicator

product & price #sales

context demand

–Katherine Barr, Partner at VC-firm MDV

"Pairing human workers with machine learning and automation

will transform knowledge work and unleash new levels of human

productivity and creativity."

Decisions from predictions

1. Descriptive

2. Predictive

3. Prescriptive

Phases of data analysis

1. Show churn rate against time

2. Predict which customers will churn next

3. Suggest what to do about each customer (e.g. propose to switch plan, send promotional offer, etc.)

Churn analysis

1. Show returned goods against {type, customer segment}

2. Predict risk shopper will return goods

E- commerce returns

“Suggest what to do about each customer” → prioritised list of actions, based on…

• Customer representation + context

• Churn prediction & action prediction

• Uncertainty in predictions

• Revenue brought by customer & Cost of actions

• Constraints on frequency of solicitations34

Churn analysis

Pric ing optimisat ion

Again, from David Jones (@d_jones, see source)

Decide price given product and context…

• For several price candidates (within constrained range):

• Predict # sales given product, context, price

• Multiply by price to estimate revenue

Decide price given product and context…

• For several price candidates (within constrained range):

• Predict 95%-confidence lower bound on # sales given product, context, price

• Multiply by price to estimate revenue

1. Show past demand against calendar

2. Predict demand for [product] at [store] in next 2 days

3. Suggest how much to ship

• Trade-off: cost of storage vs risk of lost sales

• Constraints on order size, truck volume, capacity of people putting stuff into shelves

Replenishment

• Context

• Predictions

• Uncertainty in predictions

• Constraints

• Costs / benefits

• Competing objectives (⇒ trade-offs to make)

• Business rules39

Decis ions are based on…

Who per forms better?

Star Wars: The Flat Awakens by Filipe de Carvalho

AI + Human per form better

Human alone per forms better : dex terit y

AI alone per forms better : replenishment

Decisions are faster, cheaper, and better

AI alone per forms better : replenishment

Again, from Lars Trieloff @trieloff (see source)

Decision Quality

Status Quo Predictive Prescriptive Automation

1. Descriptive analysis

2. Predictive analysis

3. Prescriptive analysis

4. Automated decisions

B eyond prescr ipt ive analysis

• Spam filter → decide to skip inbox

• Autonomous Vehicles → decide who to kill

Autonomous decis ion-mak ing systems

⇒ “Tool AI” vs “High-stakes autonomous AI”

Autonomous Vehicles

• Morality in decision-making algorithm:

• Minimize loss of life

• Account for probabilities of survival, age of occupants…→ optimal formula?

• Sacrifice owner?

• “People are in favor of cars that sacrifice the occupant to save other lives—as long they don’t have to drive one themselves.”

Autonomous Vehicles

• Need wide acceptation to get adoption and provide benefit (e.g. save lives with AVs)

• “The public is much more likely to go along with a scenario that aligns with their own views”

• What will the public tolerate? → experimental ethics

• Similar issues whenever AI decides for us and impacts many

⇒ “Domain-specific/business rules” in decision making49

H igh-stakes autonomous AIs

Role of APIs

Communication bet ween AIs

01000101101

Software components for automated decisions:

• Create training dataset from historical data (merge sources, aggregate…)

• Provide predictive model from given training set (i.e. learn)

• Provide prediction against model for given context

• Provide optimal decision from given contextual data, predictions, uncertainties, constraints, objectives, costs

• Apply given decision52

S eparation of concerns

Operations Research component

M achine Learning components

Predic t ive APIs

The two phases of machine learning:

• TRAIN a model

• PREDICT with a model

Predic t ive APIs

The two methods of predictive APIs:

• TRAIN a model

• PREDICT with a model

Predic t ive APIs

The two methods of predictive APIs:

• model = create_model(‘training.csv’)

• predicted_output = create_prediction(model, new_input)

Predic t ive APIs

Amazon ML

Google Prediction

PredicSis

Providers of REST http Predic t ive APIs

Going further

• Define desired and acceptable behaviour→ objectives and constraints/bounds

• Monitor accuracy & bottomline

• Self-monitoring & anomaly detection→ thresholds and fallbacks

Ensuring per formance of autonomous AI systems

Per formance guarantees?

“construction worker in orange safety vest is working on road”

95%-accurate scene description

Per formance guarantees

“black and white dog jumps over bar”

“a young boy is holding a baseball bat”

“a young boy is holding a baseball bat”weapon

SIR, DROP THE WEAPON!

• Lars Trieloff: “Business reasons for automating decisions”

• Daniel Kahneman: “Thinking, Fast and Slow”

• Tom Dietterich: “Artificial Intelligence Progress”

• MIT Technology Review: “Why Self-Driving Cars Must Be Programmed to Kill”

• Conference: PAPIs Connect67

Learn more

• Free ML resources: louisdorard.com

• PAPIs updates: @papisdotio

@louisdorard

LIKEDIT?THEN