liveBook · Manning

SQL

As explained in Deep Learning with R, Third Edition

A query is the representation whose information need determines which keys receive high attention relevance.

Query names the information need used to retrieve relevant source information. In attention, it is compared with keys, and the resulting scores weight the values that produce the output. section 15.3.1

Photo search

A photo search term acts as the query, photo tags act as keys, and the matching photos act as values. Attention follows the same lookup pattern. section 15.3.1

A query participates in attention through these steps:

How Deep Learning with R, Third Edition covers it

How deep it goes
Defines queries through lookup terminology and implements their projections, dot products, softmax scores, and value aggregation.
Examples
Uses photo search, target-source sequence attention, and R-style dot-product and parameterized attention listings.
What you can do after
Apply it via dot-product or parameterized attention to retrieve and combine context from source sequences.
How it connects
Connects queries to attention, keys, values, dot products, softmax, source sequences, and target sequences.
Cover of Deep Learning with R, Third Edition
liveBookmanning.com
Read Deep Learning with R, Third Edition on liveBook Get the book at manning.com

Also explained in

Compare how 25 books explain SQL →

Related concepts

This explainer was generated by Manning's knowledge graph from the text of Deep Learning with R, Third Edition. The section links point to the parts of the book it draws on.