AI Research Atlas
Journal article

Theory

The expressive power of a hidden layer

Under stated conditions, feedforward networks can approximate broad classes of functions.

Kurt Hornik, Maxwell Stinchcombe and Halbert White

AI topics

Explore related entries. Larger tags appear on more entries.

The contribution

This theoretical work established approximation capabilities of multilayer feedforward networks. It helped explain why even a network with a single hidden layer, when given enough units and suitable assumptions, could represent a rich set of input-output relationships.

What this does not establish

Representational capacity does not guarantee that training finds the right weights, that the required network is small, or that predictions generalize.

Why this date?

1989 dates this particular universal-approximation paper; it is not a claim of sole priority for the theorem family.

This entry follows the linked publication. Read the source and date conventions.

Comments

Discuss this research, ask a question, or suggest a correction. Comments appear after the site owner approves them.

Loading comments…

Sign in with ChatGPT to comment

Use your OpenAI account. Published comments show the display name you choose, not your account email.