liveBook · Manning

Signal

As explained in Machines that Think

A feature is a learned or specified attribute representing a meaningful aspect of an input for computational processing.

A feature is a representation of a meaningful distinction in input. Neural networks can learn features from examples rather than being explicitly told what to detect, building increasingly abstract representations as patterns combine. section 1.4.3 section 4.2.1

In language models, learned weights constrain how features interact and which distinctions remain separate. More parameters, wider layers, and deeper stacks increase representational capacity, but do not guarantee performance; weights encode patterns, not sentences or facts stored literally. section 4.2.1

From simple patterns to abstraction

A neural network may develop features that progress from edges to shapes and objects, or from letters to words and meaning. section 1.4.3

How Machines that Think covers it

How deep it goes
Traces features from simple patterns to abstraction and relates representational capacity to parameters and layers, without detailing individual feature vectors.
Examples
Uses edges, shapes, and objects, alongside letters, words, and meaning, to illustrate increasing abstraction.
What you can do after
Understand how data scale and model capacity support increasingly abstract feature representations without implying literal storage of facts.
How it connects
Connects features to neural networks, parameters, layers, and internal representations.

Also explained in

Compare how 14 books explain signal →

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.