Patent · US Active

Systems and methods for diversity-enhanced adaptive experimentation in a machine learning-based formulation platform

US12079697B2 · kind B2 · utility

0Cited by
0References
19Claims
0Family size

Assignee

Inventors

Key dates

Filing dateFeb 9, 2023
Grant dateSep 3, 2024
Priority date
Expiry dateFeb 9, 2043

Classification

  • Technology area (CPC G)Physics
  • CPC primaryG06N5/022
  • WIPO fieldComputer technology
  • WIPO sectorElectrical engineering

Abstract

A method and system for adaptively generating experimental product formulations for accelerating a formulation design of a target product includes constructing a formulation objective function based on a plurality of distinct product-informative formulation parameters of the target product; generating, via a formula-generating machine learning model, a distinct set of product formulation parameter values based on the formulation objective function; constructing a corpus of experimental findings data based on the distinct set of product formulation parameter values generated by the formula-generating machine learning model; automatically adapting the formula-generating machine learning model based on the corpus of experimental findings that accelerate the formulation design of the target product; and generating, via the adapted formula-generating machine learning model, an adapted set of product formulation parameter values that satisfies the one or more formulation objectives thereby enabling an accelerated creation of the target product.

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