US Patent for Strategy Learning Method Granted to Robert Bosch GmbH

ChangeBridge: Patent Grants - Business Methods (G06Q)

Published March 24th, 2026

Detected March 25th, 2026

Summary

The USPTO has granted US Patent 12585963B2 to Robert Bosch GmbH for a method and device for learning and implementing strategies, specifically for optimizing evolutionary algorithms. The patent details a method for learning a strategy that optimally adapts parameters of an evolutionary algorithm using reinforcement learning.

View original document View source feed page

What changed

The United States Patent and Trademark Office (USPTO) has granted US Patent 12585963B2 to Robert Bosch GmbH. This patent covers a method and device for learning a strategy that optimally adapts at least one parameter of an evolutionary algorithm. The method utilizes reinforcement learning, where the strategy is learned from interactions of the CMA-ES algorithm with a parameterization determined by the strategy itself, based on state information and a reward signal.

This patent grant is a routine event for a patent holder and does not impose new regulatory obligations on other entities. It represents an intellectual property right for Robert Bosch GmbH concerning their innovation in strategy learning for evolutionary algorithms. Compliance officers in related technology sectors should be aware of such patent grants as they can influence competitive landscapes and potential licensing opportunities.

Source document (simplified)

← USPTO Patent Grants

Method and device for learning a strategy and for implementing the strategy

Grant US12585963B2 Kind: B2 Mar 24, 2026

Assignee

ROBERT BOSCH GMBH

Inventors

Steven Adriaenssen, Andre Biedenkapp, Frank Hutter, Gresa Shala, Marius Lindauer, Noor Awad

Abstract

A method for learning a strategy, which optimally adapts at least one parameter of an evolutionary algorithm. The method includes the following steps: initializing the strategy, which ascertains a parameterization of the parameter as a function of pieces of state information; learning the strategy with the aid of reinforcement learning, it being learned from interactions of the CMA-ES algorithm with a parameterization, determined with the aid of the strategy as a function of the pieces of state information, with the problem instance and with a reward signal, which parameterization is optimal for possible pieces of state information.