Learning method and apparatus utilizing genetic algorithms
First Claim
1. A learning apparatus for building, based on learning data, a network structure of a dynamic Bayesian network for inferring behavior of a user, wherein in the network a cause and effect relationship between plural nodes is represented by a directed graph, the learning apparatus comprising:
- an image capturing device for observing the user,wherein the image capturing device captures one or more images of theuser;
storage means in which the learning data is stored; and
learning means for building the network structure using a K2 algorithm based on the learning data;
wherein the learning meansrecognizes one or more characteristics of the user based on one or more parameters including direction, position, size and motion of the user in one or more regions of each of the one or more images,prepares an initial population of individuals constituted by individuals each having a genotype in which orders between the plural nodes and cause and effect relationship have been stipulated,repeatedly performs processing for crossovers and/or mutations on the initial population of individuals based on a genetic algorithm,calculates an evaluated value of each of the individuals resulting from the crossovers and/or mutations based on the learning data, wherein the evaluated value of the individuals increases as generation alteration is repeated,searches for an optimum individual from the individuals resulting from the crossovers and/or mutations, andselects a phenotype of the optimum individual as the network structure.
1 Assignment
0 Petitions
Accused Products
Abstract
A learning apparatus for building a network structure of a Bayesian network based on learning data. In the Bayesian network, a cause and effect relationship between plural nodes is represented by a directed graph. The learning apparatus includes a storage portion in which the learning data is stored and a learning portion for building the network structure based on the learning data. The learning portion prepares an initial population of individuals formed by individuals each having a genotype in which orders between the nodes and cause and effect relationship have been stipulated, repeatedly performs processing for crossovers and/or mutations on the initial population of individuals based on a genetic algorithm, calculates an evaluated value of each individual based on the learning data, searches for an optimum one of the individuals, and takes a phenotype of the optimum individual as the network structure.
-
Citations
8 Claims
-
1. A learning apparatus for building, based on learning data, a network structure of a dynamic Bayesian network for inferring behavior of a user, wherein in the network a cause and effect relationship between plural nodes is represented by a directed graph, the learning apparatus comprising:
-
an image capturing device for observing the user, wherein the image capturing device captures one or more images of the user; storage means in which the learning data is stored; and learning means for building the network structure using a K2 algorithm based on the learning data; wherein the learning means recognizes one or more characteristics of the user based on one or more parameters including direction, position, size and motion of the user in one or more regions of each of the one or more images, prepares an initial population of individuals constituted by individuals each having a genotype in which orders between the plural nodes and cause and effect relationship have been stipulated, repeatedly performs processing for crossovers and/or mutations on the initial population of individuals based on a genetic algorithm, calculates an evaluated value of each of the individuals resulting from the crossovers and/or mutations based on the learning data, wherein the evaluated value of the individuals increases as generation alteration is repeated, searches for an optimum individual from the individuals resulting from the crossovers and/or mutations, and selects a phenotype of the optimum individual as the network structure. - View Dependent Claims (2, 3, 4, 5, 6)
-
-
7. A learning method of building, based on learning data and using a K2 algorithm, a network structure of a dynamic Bayesian network for inferring behavior of a user, wherein in the network a cause and effect relationship between plural nodes is represented by a directed graph, the learning method comprising the steps of:
-
capturing one or more images of the user on an image capturing device; recognizing one or more characteristics of the user based on one or more parameters including direction, position, size and motion of the user in one or more regions of each of the one or more images; preparing an initial population of individuals constituted by individuals each having a genotype in which orders between the plural nodes and cause and effect relationship have been stipulated; repeatedly performing processing for crossovers and/or mutations on the initial population of individuals based on a genetic algorithm; calculating an evaluated value of each of the individuals resulting from the crossovers and/or mutations based on the learning data, wherein the evaluated value of the individuals increases as generation alteration is repeated; searching for an optimum individual from the individuals resulting from the crossovers and/or mutations; and selecting a phenotype of the optimum individual as the network structure.
-
-
8. A learning apparatus for building, based on learning data and using a K2 algorithm, a network structure of a dynamic Bayesian network for inferring behavior of a user, wherein in the network a cause and effect relationship between plural nodes is represented by a directed graph, the learning apparatus comprising:
-
an image capturing device for observing the user, wherein the image capturing device captures one or more images of the user; a storage portion in which the learning data is stored; and a learning portion operable to build the network structure based on the learning data; wherein the learning portion recognizes one or more characteristics of the user based on one or more parameters including direction, position, size and motion of the user in one or more regions of each of the one or more images, prepares an initial population of individuals constituted by individuals each having a genotype in which orders between the plural nodes and cause and effect relationship have been stipulated, repeatedly performs processing for crossovers and/or mutations on the initial population of individuals based on a genetic algorithm, calculates an evaluated value of each of the individuals resulting from the crossovers and/or mutations based on the learning data, wherein the evaluated value of the individuals increases as generation alteration is repeated, searches for an optimum individual from the individuals resulting from the crossovers and/or mutations, and selects a phenotype of the optimum individual as the network structure.
-
Specification