This document provides an overview of machine learning and artificial intelligence. It discusses how machine learning works by having algorithms learn from large amounts of data to make decisions without being explicitly programmed. The document outlines the step-by-step machine learning process of defining an objective, preparing data, training a solution using an algorithm, and testing the model. It also discusses how AI can be used for tasks like vision, communication, and games. Examples of real-world AI applications are also provided.
18. Confidential & Proprietary
I’d like:
Y/N with at least 80% accuracy.
Step 1: Define your objective
?
(information)
Y / N
(answer)
?
(recipe)
@quaesita
29. Confidential & Proprietary
Train your solution
?
(information)
Y / N
(answer)
Pick an algorithm!
?
(recipe)
Age
Review Score
(information)
@quaesita
30. Confidential & Proprietary
Train your solution
Wine classification
Age in years
Reviewscore
4
2
0
-2
2 4 6 8
Support Vector Classifier
Decision Tree
Neural Network
Algorithm selection:
@quaesita
31. Confidential & Proprietary
Wine classification
Age in years
Reviewscore
4
2
0
-2
2 4 6 8
Support Vector Classifier
Decision Tree
Neural Network
Algorithm selection:
Train your solution
@quaesita
32. Confidential & Proprietary
Support Vector Classifier
Algorithm selected:
Wine classification
Age in years
Reviewscore
4
2
0
-2
2 4 6 8
Run ML algorithm
Train your solution
@quaesita
33. Confidential & Proprietary
Support Vector Classifier
Algorithm selected:
Run ML algorithm
Wine classification
Age in years
Reviewscore
4
2
0
-2
2 4 6 8
Train your solution
@quaesita
34. Confidential & Proprietary
Get model (recipe)
Wine classification
Age in years
Reviewscore
4
2
0
-2
2 4 6 8
Train your solution
@quaesita
38. Confidential & Proprietary
Support Vector Classifier
Decision Tree
Neural Network
Algorithm selection:
Wine classification
Age in years
Reviewscore
4
2
0
-2
2 4 6 8
Train your solution
@quaesita
39. Confidential & Proprietary
Support Vector Classifier
Decision Tree
Neural Network
Algorithm selection:
Wine classification
Age in years
Reviewscore
4
2
0
-2
2 4 6 8
Train your solution
@quaesita
40. Confidential & Proprietary
Run ML algorithm
Wine classification
Age in years
Reviewscore
4
2
0
-2
2 4 6 8
Decision Tree
Algorithm selected:
Train your solution
@quaesita
41. Confidential & Proprietary
Decision Tree
Algorithm selected:
Run ML algorithm
Wine classification
Age in years
Reviewscore
4
2
0
-2
2 4 6 8
Train your solution
@quaesita
42. Confidential & Proprietary
Decision Tree
Algorithm selected:
Run ML algorithm
Wine classification
Age in years
Reviewscore
4
2
0
-2
2 4 6 8
Train your solution
@quaesita
43. Confidential & Proprietary
Get model (recipe)
Wine classification
Age in years
Reviewscore
4
2
0
-2
2 4 6 8
Train your solution
@quaesita
56. Confidential & Proprietary@quaesita
And the winner is...
Statistical Output (CI): [80%, 100%] Statistical Output (CI): [51%, 90%]
Support vector classifier Decision tree
57. Confidential & Proprietary
Testing is crucial!
It’s the key to responsible ML and AI.
Make sure your solution actually works
on relevant, new data.
@quaesita
58. Confidential & Proprietary
Applied ML is easier than
most people think
Tinker and have fun, try out anything you like,
but assess performance carefully.
@quaesita
60. Confidential & Proprietary
Machine learning is
an approach to making repeated
decisions that involves algorithmically
finding patterns in data and using
these to make recipes that deal
correctly with brand new data.
@quaesita
74. Confidential & Proprietary 74Proprietary + Confidential
GPUs
Researchers began to notice that
neural network mathematics closely
resembled the algorithms to shade
pixels in graphics cards (GPUs)
@quaesita
87. Confidential & Proprietary
• Predictive inventory planning
• Recommendation engines
• Upsell and cross-channel marketing
• Market segmentation and targeting
• Customer ROI and lifetime value
Retail
• Alerts and diagnostics from real-time
patient data
• Disease identification and risk stratification
• Patient triage optimization
• Proactive health management
• Healthcare provider sentiment analysis
Healthcare and Life Sciences
• Aircraft scheduling
• Dynamic pricing
• Social media – consumer feedback
and interaction analysis
• Customer complaint resolution
• Traffic patterns and congestion
management
Travel and Hospitality
• Risk analytics and regulation
• Customer Segmentation
• Cross-selling and up-selling
• Sales and marketing campaign
management
• Credit worthiness evaluation
Financial Services
• Predictive maintenance or
condition monitoring
• Warranty reserve estimation
• Propensity to buy
• Demand forecasting
• Process optimization
• Telematics
Manufacturing
• Power usage analytics
• Seismic data processing
• Carbon emissions and trading
• Customer-specific pricing
• Smart grid management
• Energy demand and supply
optimization
Energy, Feedstock and
Utilities
@quaesita
88. ML PlatformML Pre-Trained APIs
Cloud
Vision
Cloud Natural
Language
Cloud
Translation
Cloud
Speech
Cloud Video
Intelligence
ML Accelerators
Cloud GPU Cloud TPU
Cloud ML
Engine
Cloud
Dataproc
Cloud
Dataflow
Google
BigQuery
Cloud Job
Discovery
ML Professional
Services & PartnersML & Data Science Tools
ML Libraries
Cloud
Datalab
Cloud
Dataprep
ASL
Google Cloud AI
@quaesita