Artificial intelligence (AI)
Artificial intelligence extends beyond what machine learning can do. It is about making decisions and aiming for success rather than accuracy. One way to think of it is that machine learning aims to gain knowledge while artificial intelligence aims for wisdom or intelligence. An example of AI in action would be Boston Dynamic's Atlas robot, which can navigate freely in the open world and avoid obstacles without the aid of human control. The robot does not fully depend on the historical map data to navigate. However, for machine learning, it's about creating or predicting a pattern from historical data analysis. Similar to the robot's navigation, it is about understanding the most optimal route by creating patterns based on historical and crowd-sourced traffic data.
Setting up a modern data warehouse with cloud analytics is the key factor in preparing to execute ML/AI. Without migrating the workloads to the cloud, deriving ML/AI models will mean encountering various roadblocks in order to maximize the business value of these emerging technologies. A modern data warehouse and analytics pipeline form the backbone that enables you to pass these roadblocks.
Microsoft is a leader in machine learning and artificial intelligence as they have been driving a lot of innovation throughout their products and tools—for instance, Window's digital assistant, Cortana, and Office 365's live captions and subtitles. They offer a range of products, tools, and services such as Microsoft Cognitive Services, Azure Machine Learning studio, the Azure Machine Learning service, and ML.NET.
Microsoft is setting an example with their AI for Good initiative, which aims to make the world more sustainable and accessible through AI. One particularly interesting project is AI for Snow Leopards, in which Microsoft uses AI technology to detect snow leopards (who are almost invisible in snow) in order to protect the endangered species. Exploring artificial intelligence and deep learning (the ability to learn without human supervision), specifically the data science and formula aspects, is not the focus of this book, but you will tackle some concepts in later chapters (see more on this in Chapter 3, Processing and visualizing data).