Читать книгу Smarter Data Science - Cole Stryker - Страница 23

The variety of opportunities to apply machine learning is extensive. The sheer variety gives credence as to why so many different modes of learning are necessary:

 Advertisement serving

 Business analytics

 Call centers

 Computer vision

 Companionship

 Creating prose

 Cybersecurity

 Ecommerce

 Education

 Finance, algorithmic trading

 Finance, asset allocation

 First responder rescue operations

 Fraud detection

 Law

 Housekeeping

 Elderly care

 Manufacturing

 Mathematical theorems

 Medicine/surgery

 Military

 Music composition

 National security

 Natural language understanding

 Personalization

 Policing

 Political

 Recommendation engines

 Robotics, consumer

 Robotics, industry

 Robotics, military

 Robotics, outer space

 Route planning

 Scientific discovery

 Search

 Smart homes

 Speech recognition

 Translation

 Unmanned aerial vehicles (drones, cars, ambulance, trains, ships, submarines, planes, etc.)

 Virtual assistants

Evaluating how well a model learned can follow a five-point rubric.

 Phenomenal: It's not possible to do any better.

 Crazy good: Outcomes are better than what any individual could achieve.

 Super-human: Outcomes are better than what most people could achieve.

 Par-human: Outcomes are comparable to what most people could achieve.

 Sub-human: Outcomes are less than what most people could achieve.