liveBook · Manning

Distance

As explained in Machine Learning System Design

A function that assigns a numerical measure of separation or dissimilarity between mathematical objects or representations.

A distance function can help combine retrieval signals into a unified relevance score. A baseline sums distances; a weighted sum offers another option. section 11.5.2

Combining photo-search signals

A photo search system can combine query-to-image and query-to-description distances with user and document features. A weighted sum amounts to training a small linear model on the distances. section 11.5.2

How Machine Learning System Design covers it

How deep it goes
Contrasts a simple sum with a weighted sum, but does not detail how to choose or train the weights.
Examples
Combines query-to-image and query-to-description distances with user and document features in a photo-search score.
What you can do after
Evaluate whether a simple sum or weighted sum should combine retrieval distances and other features.
How it connects
Connects distance aggregation with feature engineering, relevance scoring, and linear models.

Also explained in

Compare how 10 books explain distance →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Machine Learning System Design. The section links point to the parts of the book it draws on.