Processing

Please wait...

Settings

Settings

Goto Application

1. WO2022164688 - TWO-STAGE SAMPLING FOR ACCELERATED DEFORMULATION GENERATION

Publication Number WO/2022/164688
Publication Date 04.08.2022
International Application No. PCT/US2022/012888
International Filing Date 19.01.2022
IPC
G16C 20/70 2019.1
GPHYSICS
16INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
CCOMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
20Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
70Machine learning, data mining or chemometrics
G16C 20/30 2019.1
GPHYSICS
16INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
CCOMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
20Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
30Prediction of properties of chemical compounds, compositions or mixtures
G16C 20/20 2019.1
GPHYSICS
16INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
CCOMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
20Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
20Identification of molecular entities, parts thereof or of chemical compositions
G16C 20/00 2019.1
GPHYSICS
16INFORMATION AND COMMUNICATION TECHNOLOGY SPECIALLY ADAPTED FOR SPECIFIC APPLICATION FIELDS
CCOMPUTATIONAL CHEMISTRY; CHEMOINFORMATICS; COMPUTATIONAL MATERIALS SCIENCE
20Chemoinformatics, i.e. ICT specially adapted for the handling of physicochemical or structural data of chemical particles, elements, compounds or mixtures
CPC
A23L 5/00
AHUMAN NECESSITIES
23FOODS OR FOODSTUFFS; TREATMENT THEREOF, NOT COVERED BY OTHER CLASSES
LFOODS, FOODSTUFFS, OR NON-ALCOHOLIC BEVERAGES, NOT COVERED BY SUBCLASSES A23B - A23J; THEIR PREPARATION OR TREATMENT, e.g. COOKING, MODIFICATION OF NUTRITIVE QUALITIES, PHYSICAL TREATMENT
5Preparation or treatment of foods or foodstuffs, in general; Food or foodstuffs obtained thereby; Materials therefor
A23V 2002/00
AHUMAN NECESSITIES
23FOODS OR FOODSTUFFS; TREATMENT THEREOF, NOT COVERED BY OTHER CLASSES
VINDEXING SCHEME RELATING TO FOODS, FOODSTUFFS OR NON-ALCOHOLIC BEVERAGES
2002Food compositions, function of food ingredients or processes for food or foodstuffs
G06F 18/2155
G06F 18/22
G06F 18/24
G06F 18/2415
Applicants
  • CITRINE INFORMATICS, INC. [US]/[US]
Inventors
  • SEVGEN, Selami, Emre
  • FOLIE, Brendan, David
  • LING, Julia, Black
Agents
  • BROWNSTONE, Daniel, R.
  • AHN, Dohyun
  • FARN, Michael, W.
  • HULSE, Robert, A.
  • HASSAN, Saad, K.
Priority Data
63/141,72326.01.2021US
Publication Language English (en)
Filing Language English (EN)
Designated States
Title
(EN) TWO-STAGE SAMPLING FOR ACCELERATED DEFORMULATION GENERATION
(FR) ÉCHANTILLONNAGE EN DEUX ÉTAPES POUR LA GÉNÉRATION DE DÉFORMULATION ACCÉLÉRÉE
Abstract
(EN) A device receives an ingredient list having a sequence of ingredients ordered by relative amount, and generates formulation vectors by sampling the ingredients list. The device inputs the plurality of formulation vectors into a machine-learned model, the machine-learned model generating an encoded version of each of the plurality of formulation vectors using an encoder, and then outputting a plurality of reconstructed formulation vectors as derived using a decoder. The device identifies reconstructed formulation vectors that have an order that matches the sequence, defines a latent space using the encoded version of the matching reconstructed formulation vectors. The device iteratively samples the latent space until a threshold number of samples are derived that match an ordering constraint that corresponds to the sequence, performs a statistical aggregation of the samples, and outputs an indication of an absolute amount of each ingredient in the ingredients list.
(FR) Selon la présente invention, un dispositif reçoit une liste d'ingrédients ayant une séquence d'ingrédients ordonnée par quantité relative, et génère des vecteurs de formulation par échantillonnage de la liste d'ingrédients. Le dispositif entre la pluralité de vecteurs de formulation dans un modèle appris par machine, le modèle appris par machine générant une version codée de chacun de la pluralité de vecteurs de formulation à l'aide d'un encodeur, puis délivrant une pluralité de vecteurs de formulation reconstruits tels que dérivés à l'aide d'un décodeur. Le dispositif identifie des vecteurs de formulation reconstruits qui ont un ordre qui correspond à la séquence, définit un espace latent à l'aide de la version codée des vecteurs de formulation reconstruits qui correspondent. Le dispositif échantillonne de manière itérative l'espace latent jusqu'à ce qu'un nombre seuil d'échantillons soient dérivés qui correspondent à une contrainte de commande correspondant à la séquence, effectue une agrégation statistique des échantillons, et délivre une indication d'une quantité absolue de chaque ingrédient dans la liste d'ingrédients.
Latest bibliographic data on file with the International Bureau