Looking for breakthrough ideas for innovation challenges? Try Patsnap Eureka!

Information processing device, information processing method, and program

An information processing device and time series technology, applied in the field of learning models, can solve problems such as difficult distinction and perceptual confusion

Inactive Publication Date: 2011-04-13
SONY CORP
View PDF3 Cites 9 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0008] In addition, since state transition information is not obtained, it is not easy to deal with a problem called perceptual confusion, that is, a problem in which the sensor signals to be observed from the modeled object are the same even when the states of the modeled object are different. In the case of , it is not easy to distinguish that
[0032] For the description in R.B. Chinnam, P.Baruah, "Autonomous Diagnostics and Prognostics Through Competitive Learning Driven HMM-BasedClustering", Proceedings of the International Joint Conference on NeuralNetworks, July 20-24, 2003, Vol. 4, pp. 2466-2471 The method is also related to that in Panu Somervuo, "Competing Hidden Markov Models on the Self-Organizing Map", ijcnn, p. 3169, IEEE-INNS-ENNS International Joint Conference on Neural Networks (IJCNN′00) - Volume 3, 2000 In the same way as the method described in , the number of modules must be set in advance, and it is difficult to apply it to the situation where the scale of the modeled object is not known in advance

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Information processing device, information processing method, and program
  • Information processing device, information processing method, and program
  • Information processing device, information processing method, and program

Examples

Experimental program
Comparison scheme
Effect test

no. 1 example

[0109] Configuration examples for learning devices

[0110] figure 1 is a block diagram showing a configuration example of the first embodiment of the learning device to which the information processing device according to the present invention is applied.

[0111] exist figure 1 In , the learning device learns a learning model (executes modeling) for providing statistical dynamic properties of the modeled object based on observation values ​​to be observed from the modeled object.

[0112] Now, it is assumed that the learning device has no preliminary knowledge about the modeled object, but may have preliminary knowledge.

[0113] The learning device includes a sensor 11 , an observation time series buffer 12 , a module learning unit 13 , an identification unit 14 , a transition information management unit 15 , an ACHMM (Additional Competition Hidden Markov Model) storage unit 16 and an HMM configuration unit 17 .

[0114] The sensor 11 senses the modeling object at each t...

no. 2 example

[0656] As described above, by applying ACHMM to an actor who performs actions autonomously, ACHMM learning is performed at the actor using the time series of observations to be observed from the exercise environment, so that a map of the exercise environment can be obtained by ACHMM.

[0657] Furthermore, for the actor, the combined HMM is reconfigured according to ACHMM, and the plan is obtained using the combined HMM, i.e., from the current state The maximum likelihood value state sequence to the goal state #g, according to the plan to execute the action, so that the actor can move from the current state in the motion environment The corresponding position is moved to the position corresponding to the goal state #g.

[0658] Incidentally, for the combined HMM reconfigured according to the ACHMM, state transitions that are not actually realized can be expressed as if realized probabilistically.

[0659] specifically, Figure 35 is a diagram showing an example of reconfigu...

no. 3 example

[0977] Figure 58 is described by Figure 8 A flowchart of another example of the module learning process performed by the module learning unit 13 in .

[0978] Note that for Figure 58 The modules in the learning process, execute Figure 17 variable window learning as described in , but it is also possible to perform Figure 9 Fixed-window learning as described in .

[0979] for Figure 9 with Figure 17 Modules in the learning process, such as Figure 10 As described in , according to the maximum log likelihood value maxLP (that is, the maximum likelihood value module #m ＊ The logarithmic likelihood value of ) and the size correlation between the predetermined threshold likelihood value TH, the maximum likelihood value module #m ＊ Or the new module is identified as the object module.

[0980] Specifically, in the case where the maximum log likelihood value maxLP is equal to or greater than the threshold likelihood value TH, the maximum likelihood value module #m ＊ ...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

An information processing device, an information processing method, and a program. The information processing device comprising: a likelihood calculating unit configured to take the time series of an observed value to be successively supplied as learned data to be used for learning, and with regard to each module making up a learning model having an HMM (Hidden Markov Model) as a module which is the minimum component, to obtain likelihood that the learned data may be observed at the module; an object module determining unit configured to determine, based on the likelihood, a single module of the learning model, or a new module to be an object module that is an object module having an HMM parameter to be updated; and an updating unit configured to perform learning for updating the HMM parameter of the object module using the learned data.

Description

technical field [0001] The present invention relates to an information processing device, an information processing method, and a program, and more specifically, the present invention relates to an information processing device, an information processing method, and a program that enable a learning model of an appropriate scale to be obtained for a modeling object. Background technique [0002] A method for sensing a modeled object (an object to be modeled by a sensor) and modeling a sensor signal to be output by the sensor using an observation value (learning of a learning model) includes a method for modeling the sensor signal (observation value) The k-means clustering method and SOM (self-organizing map) for clustering. [0003] For example, with the k-means clustering method and SOM, if we consider a certain state (internal state) of the modeled object to correspond to a cluster, then set the state within the signal space of the sensor signal (the observation space of th...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06N7/00G06N20/00G06N20/10
CPCG06N99/005G06N20/00G06N20/10
Inventor 铃木洋贵
Owner SONY CORP
Who we serve
  • R&D Engineer
  • R&D Manager
  • IP Professional
Why Patsnap Eureka
  • Industry Leading Data Capabilities
  • Powerful AI technology
  • Patent DNA Extraction
Social media
Patsnap Eureka Blog
Learn More
PatSnap group products

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2024 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com