Patent · US Active

Training a function to respond predictably to differences

US11804034B2 · kind B2 · utility

0Cited by

1References

13Claims

0Family size

Assignee

Robert Bosch GmbH · DE

Inventors

Thomas Keller · Petersburg, US
Anna Khoreva · Stuttgart, DE
Max Welling · Bussum, NL

Key dates

Filing date	Apr 16, 2021
Grant date	Oct 31, 2023
Priority date	—
Expiry date	Jan 6, 2042

Classification

Technology area (CPC G)Physics
CPC primaryG06V20/70
WIPO fieldComputer technology
WIPO sectorElectrical engineering

Abstract

A computer-implemented method of training a machine learnable function, such as an image classifier or image feature extractor. When applying such machine learnable functions in autonomous driving and similar application areas, generalizability may be important. To improve generalizability, the machine learnable function is rewarded for responding predictably at a layer of the machine learnable function to a set of differences between input observations. This is done by means of a regularization objective included in the objective function used to train the machine learnable function. The regularization objective rewards a mutual statistical dependence between representations of input observations at the given layer, given a difference label indicating a difference between the input observations.

Source: USPTO / EPO open patent data. Objective bibliographic and citation counts.