US 2017/0372197 A1Published application

Customized cooking utilizing deep learning neuromorphic computing of hyperspectral input

AssigneeInternational Business Machines Corporation

InventorsAaron K. Baughman, Rick A. Hamilton, II, Sathya Santhar, Ashish K. Tanuku

Life of this patent

Estimating the term…

Loading patent…

Abstract

One or more processors receive hyperspectral band input, biometric input, and cognitive input as response input, from a user sampling a plurality of base foods, each base food prepared with a subset of ingredients and preparation techniques. The response input is transformed to a numeric representation of the respective input. Deep learning techniques are used to train an algorithm using the response data. A probabilistic ranking of base food is generated using unsupervised learning. Probability values of base food, ingredients, and preparation technique, associations preferred by the user, are generated, along with rules which define constraints associated with conditions for base food, ingredient, and preparation techniques, of user preferences. An objective function is generated that includes decision variables respectively aligned with constraints, and in response to optimizing the objective function, a preferred base food and ingredients, with preferred conditions of the user, is determined.

Claims20

Dependent claims

A method comprising:

receiving, by one or more processors, somatic and cognitive response data of a user sampling a plurality of base foods, wherein each instance of a sample of a base food is prepared with a set of ingredients and preparation techniques from a plurality of ingredients and preparation techniques, and wherein the somatic and the cognitive response data includes labeled and unlabeled data;
determining, by one or more processors, hierarchical features of the somatic and the cognitive response data utilizing unsupervised deep learning techniques of machine learning performed on the unlabeled data of the somatic and the cognitive response data that is received;
generating by one or more processors, a probabilistic ranking of the user's preference of the plurality of base foods, based on results of the unsupervised deep learning techniques of machine learning;
performing an association, by one or more processors, of one or more sets of ingredients and preparation techniques from the plurality of ingredients and preparation techniques, with one or more base foods of the plurality of base foods, respectively, based on supervised deep learning techniques of machine learning, using the somatic and the cognitive response data of responses that are labeled;
generating, by one or more processors, a probability of the user preferring a base food of the plurality of base foods associated with a set of ingredients and preparation techniques of the plurality of ingredients and preparation techniques;
generating by one or more processors, an objective function that includes a decision variable corresponding respectively to an element of the set of ingredients and preparation techniques;
and
in response to optimizing the objective function, one or more processors determining an amount of the element of the set of ingredients and preparation techniques to include in preparation of the base food preferred by the user.

5 more elements

A computer program product, comprising:

one or more computer-readable storage media having program instruction executable by one or more computer processors, the program instructions comprising:
program instructions to receive somatic and cognitive response data of a user sampling a plurality of base foods, wherein each instance of a sample of a base food is prepared with a set of ingredients and preparation techniques from a plurality of ingredients and preparation techniques, and wherein the somatic and the cognitive response data includes labeled and unlabeled data;
program instructions to determine hierarchical features of the somatic and the cognitive response data utilizing unsupervised deep learning techniques of machine learning performed on the unlabeled data of the somatic and the cognitive response data that is received;
program instructions to generate a probabilistic ranking of the user's preference of the plurality of base foods, based on results of the unsupervised deep learning techniques of machine learning;
program instructions to perform an association of one or more sets of ingredients and preparation techniques from the plurality of ingredients and preparation techniques, with one or more base foods of the plurality of base foods, respectively, based on supervised deep learning techniques of machine learning, using the somatic and the cognitive response data of responses that are labeled;
program instructions to generate a probability of the user preferring a base food of the plurality of base foods associated with a set of ingredients and preparation techniques of the plurality of ingredients and preparation techniques;
program instructions to generate an objective function that includes a decision variable corresponding respectively to an element of the set of ingredients and preparation techniques;
and
in response to optimizing the objective function, program instructions to determine an amount of the element of the set of ingredients and preparation techniques to include in preparation of the base food preferred by the user.

6 more elements

A system comprising:

one or more computer processors, one or more computer-readable storage media, program instructions stored on the one or more computer-readable storage media for execution by at least one of the one or more processors, the program instructions comprising:
program instructions to receive somatic and cognitive response data of a user sampling a plurality of base foods, wherein each instance of a sample of a base food is prepared with a set of ingredients and preparation techniques from a plurality of ingredients and preparation techniques, and wherein the somatic and the cognitive response data includes labeled and unlabeled data;
program instructions to determine hierarchical features of the somatic and the cognitive response data utilizing unsupervised deep learning techniques of machine learning performed on the unlabeled data of the somatic and the cognitive response data that is received;
program instructions to generate a probabilistic ranking of the user's preference of the plurality of base foods, based on results of the unsupervised deep learning techniques of machine learning;
program instructions to perform an association of one or more sets of ingredients and preparation techniques from the plurality of ingredients and preparation techniques, with one or more base foods of the plurality of base foods, respectively, based on supervised deep learning techniques of machine learning, using the somatic and the cognitive response data of responses that are labeled;
program instructions to generate a probability of the user preferring a base food of the plurality of base foods associated with a set of ingredients and preparation techniques of the plurality of ingredients and preparation techniques;
program instructions to generate an objective function that includes a decision variable corresponding respectively to an element of the set of ingredients and preparation techniques;
and
in response to optimizing the objective function, program instructions to determine an amount of the element of the set of ingredients and preparation techniques to include in preparation of the base food preferred by the user.

6 more elements

Dependencies are inferred from the claim text.

Facts

Application: 15/193,136Filed 27 Jun 2016
Published: 28 Dec 2017
Term ends: Not estimatedThe source records don’t support an estimate.

Classification14 codes

A23L 33/30
Modifying nutritive qualities of foods; Dietetic products; Preparation or treatment thereof: dietetic or nutritional methods, e.g. for losing weight
A23L 5/00
Preparation or treatment of foods or foodstuffs, in general
G06N 3/042
Architecture, e.g. interconnection topology: knowledge-based neural networks; Logical representations of neural networks
G06N 3/044
Architecture, e.g. interconnection topology: recurrent networks, e.g. Hopfield networks