Back to results
US 2007/0185825 A1Published application
Learning system and method, recognition device and method, creation device and method, recognition and creation device and method, and program
AssigneeSony Corporation
InventorsMasato Ito, Katsuki Minamino, Yukiko Yoshiike, Hirotaka Suzuki, Kenta Kawamoto
Life of this patent
Estimating the term…
Abstract
A learning system is provided, which includes network storage means for storing a network including a plurality of nodes, each of which holds a dynamics; and learning means for self-organizationally updating the dynamics of the network on the basis of measured time-series data.
Claims30
A learning system comprising:
- network storage means for storing a network including a plurality of nodes, each of which holds a dynamics;
- and
- learning means for self-organizationally updating the dynamics of the network on the basis of measured time-series data.
A recognition device comprising:
- network storage means for storing a network including a plurality of nodes, in which a dynamical approximate model having an inner state quantity is set as a node and a dynamics is held in each node;
- and
- recognition means for determining a winner node which is a node holding the dynamics most suitable for measured time-series data while updating the inner state quantity on the basis of the measured time-series data, and outputting information indicating the winner node as a recognition result of the measured time-series data.
A creation device comprising:
- network storage means for storing a network including a plurality of nodes, in which a dynamical approximate model having an inner state quantity is set as a node and a dynamics is held in each node;
- and
- creation means for determining a creation node which is a node used to create time-series data and creating the time-series data while updating the inner state quantity of the creation node.
A recognition and creation device comprising:
- network storage means for storing a network including a plurality of nodes, in which a dynamical approximate model having an inner state quantity is set as a node and a dynamics is held in each node;
- recognition means for determining a winner node which is a node holding the dynamics most suitable for measured time-series data while updating the inner state quantity on the basis of the measured time-series data, and outputting information indicating the winner node as a recognition result of the measured time-series data;
- inner state storage means for storing the inner state quantity acquired in determining the winner node;
- and
- creation means for determining the winner node as a creation node which is a node used to create time-series data on the basis of the recognition result, determining the inner state quantity of the creation node on the basis of the value stored in the inner state storage means, and creating the time-series data while updating the inner state quantity of the creation node.
2 more elements
A learning method comprising the step of:
- self-organizationally updating dynamics of a network including a plurality of nodes, each of which holds a dynamics, on the basis of measured time-series data.
A program having a computer execute a learning process comprising the step of:
- self-organizationally updating dynamics of a network including a plurality of nodes, each of which holds a dynamics, on the basis of measured time-series data.
A recognition method comprising the step of:
- determining a winner node which is a node holding the dynamics most suitable for measured time-series data while updating the inner state quantity on the basis of the measured time-series data, among nodes of a network including a plurality of nodes in which a dynamical approximate model having an inner state quantity is set as a node and a dynamics is held in each node, and outputting information indicating the winner node as a recognition result of the measured time-series data.
A program having a computer execute a recognition process comprising the step of:
- determining a winner node which is a node holding the dynamics most suitable for measured time-series data while updating the inner state quantity on the basis of the measured time-series data, among nodes of a network including a plurality of nodes in which a dynamical approximate model having an inner state quantity is set as a node and a dynamics is held in each node, and outputting information indicating the winner node as a recognition result of the measured time-series data.
A creation method comprising the step of:
- determining a creation node which is a node used to create time-series data, among nodes of a network including a plurality of nodes in which a dynamical approximate model having an inner state quantity is set as a node and a dynamics is held in each node, and creating the time-series data while updating the inner state quantity of the creation node.
A program having a computer execute a creation process comprising the step of:
- determining a creation node which is a node used to create time-series data, among nodes of a network including a plurality of nodes in which a dynamical approximate model having an inner state quantity is set as a node and a dynamics is held in each node, and creating the time-series data while updating the inner state quantity of the creation node.
A recognition and creation method comprising the steps of:
- determining a winner node which is a node holding the dynamics most suitable for measured time-series data while updating the inner state quantity on the basis of the measured time-series data, among nodes of a network including a plurality of nodes in which a dynamical approximate model having an inner state quantity is set as a node and a dynamics is held in each node, and outputting information indicating the winner node as a recognition result of the measured time-series data;
- storing the inner state quantity acquired in determining the winner node in inner state storage means;
- and
- determining the winner node as a creation node which is a node used to create time-series data on the basis of the recognition result, determining the inner state quantity of the creation node on the basis of the value stored in the inner state storage means, and creating the time-series data while updating the inner state quantity of the creation node.
1 more element
A program having a computer execute a recognition and creation process comprising the steps of:
- determining a winner node which is a node holding the dynamics most suitable for measured time-series data while updating the inner state quantity on the basis of the measured time-series data, among nodes of a network including a plurality of nodes in which a dynamical approximate model having an inner state quantity is set as a node and a dynamics is held in each node, and outputting information indicating the winner node as a recognition result of the measured time-series data;
- storing the inner state quantity acquired in determining the winner node in inner state storage means;
- and
- determining the winner node as a creation node which is a node used to create time-series data on the basis of the recognition result, determining the inner state quantity of the creation node on the basis of the value stored in the inner state storage means, and creating the time-series data while updating the inner state quantity of the creation node.
1 more element
A learning system comprising:
- a network storage unit that stores a network including a plurality of nodes, each of which holds a dynamics;
- and
- a learning unit self-organizationally updating the dynamics of the network on the basis of measured time-series data.
A recognition device comprising:
- a network storage unit that stores a network including a plurality of nodes, in which a dynamical approximate model having an inner state quantity is set as a node and a dynamics is held in each node;
- and
- a recognition unit that determines a winner node which is a node holding the dynamics most suitable for measured time-series data while updating the inner state quantity on the basis of the measured time-series data, and outputs information indicating the winner node as a recognition result of the measured time-series data.
A creation device comprising:
- a network storage unit that stores a network including a plurality of nodes, in which a dynamical approximate model having an inner state quantity is set as a node and a dynamics is held in each node;
- and
- a creation unit that determines a creation node which is a node used to create time-series data and creates the time-series data while updating the inner state quantity of the creation node.
A recognition and creation device comprising:
- a network storage unit that stores a network including a plurality of nodes, in which a dynamical approximate model having an inner state quantity is set as a node and a dynamics is held in each node;
- a recognition unit that determines a winner node which is a node holding the dynamics most suitable for measured time-series data while updating the inner state quantity on the basis of the measured time-series data, and outputs information indicating the winner node as a recognition result of the measured time-series data;
- an inner state storage unit that stores the inner state quantity acquired in determining the winner node;
- and
- a creation unit that determines the winner node as a creation node which is a node used to create time-series data on the basis of the recognition result, determines the inner state quantity of the creation node on the basis of the value stored in the inner state storage unit, and creates the time-series data while updating the inner state quantity of the creation node.
2 more elements
A recognition and creation method comprising the steps of:
- determining a winner node which is a node holding the dynamics most suitable for measured time-series data while updating the inner state quantity on the basis of the measured time-series data, among nodes of a network including a plurality of nodes in which a dynamical approximate model having an inner state quantity is set as a node and a dynamics is held in each node, and outputting information indicating the winner node as a recognition result of the measured time-series data;
- storing the inner state quantity acquired in determining the winner node in an inner state storage unit;
- and
- determining the winner node as a creation node which is a node used to create time-series data on the basis of the recognition result, determining the inner state quantity of the creation node on the basis of the value stored in the inner state storage unit, and creating the time-series data while updating the inner state quantity of the creation node.
1 more element
Dependencies are inferred from the claim text.