/Patent Tracker
GoogleGranted patentAI & softwareInterest 6/10

Google patents system that saves AI predictions to speed up searches

Google has been granted a patent for a method that creates a searchable index using rules from machine learning models. The system stores previous AI predictions and their probabilities, allowing it to quickly find and reuse relevant results when similar new data is received instead of recalculating them from scratch.

Why it mattersThis technique could make AI-powered search and recommendation systems faster and more efficient by reusing past calculations for similar queries.

In the filing’s own words

Systems and techniques are disclosed for generating entries for a searchable index based on rules generated by one or more machine-learned models. The index entries can include one or more tokens correlated with an outcome and an outcome probability. A subset of tokens can be identified based on the characteristics of an event. The index may be searched for outcomes and their respective probabilities that correspond to tokens that are similar to or match the subset of tokens based on the event.

What exactly they claim (claim 1)

1. A computer-implemented method, comprising: before receiving a subsequent model input that is associated with at least one or more features: generating, using a machine learning model, data indicating a model output for a model input that is associated with the one or more features, wherein the model output is generated by the machine learning model based at least on the one or more features, and storing, in a record entry of a database, data that associates the model input with the data indicating the model output generated by the machine learning model; receiving the subsequent model input that is associated with at least one or more features; and in response to receiving the subsequent model input: accessing the record entry of the database to identify the model output that was previously generated by the machine learning model as relevant to the subsequent model input, and providing data corresponding to the model output in response to the subsequent model input.

Facts

Company
Google (Google LLC)
Official title
Searchable index
Patent number
US12730806B2
Application number
19081319
Filed
Mar 17, 2025
Granted
Sep 8, 2026
Inventors
3
Claims
19

Classified under

G06F 16/245 Physics & computingG06F 16/2228 Physics & computingG06F 16/319 Physics & computingG06N 5/025 Physics & computingG06N 20/00 Physics & computing
Read the full filing (official PDF) ↗ Open in the Patent Tracker

More from Google in AI & software

Summaries are simplified from the filing’s own abstract and first claim. A patent filing is not a product announcement.
Patent data and drawings: U.S. Patent and Trademark Office (USPTO). Nebula is not affiliated with or endorsed by the USPTO.