liveBook · Manning

Machine learning

As explained in Machines that Think

A field of artificial intelligence in which systems improve their predictions or generated outputs by learning patterns from data rather than relying only on explicitly programmed rules.

Machine learning lets systems respond to uncertainty without requiring explicit rules for every situation. Rather than demanding certainty, they weigh possible outcomes using past experience. section 1.4.2

By repeatedly encountering data, a system adjusts its internal settings to find patterns and improve predictions or generated outputs. This can help it weigh options, assess risk, and make decisions in complex environments. section 1.4.2

Machine learning is not the same as human learning. Artificial systems typically optimize externally defined objectives and separate training from inference; neural networks can learn statistical regularities without explicitly modeling what their inputs describe. Similar behavior does not establish similar mechanisms. section 5.5

How Machines that Think covers it

How deep it goes
Defines learning as updating internal settings from data and probabilities, but gives no algorithmic or mathematical detail.
What you can do after
Understand how data-driven prediction differs from explicit rules and from human learning mechanisms.
How it connects
Links machine learning to probabilistic reasoning, artificial intelligence, neural networks, objective functions, training, and inference.

Also explained in

Compare how 49 books explain machine learning →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Machines that Think. The section links point to the parts of the book it draws on.