IEICE Technical Committee Submission System
Conference Paper's Information
Online Proceedings
[Sign in]
... (for ESS/CS/ES/ISS)
Tech. Rep. Archives
... (for ES/CS)
 Go Top Page Go Previous   [Japanese] / [English] 

Paper Abstract and Keywords
Presentation 2014-03-18 16:00
Causality Trace -- Effective Retrospective Learning by Introducing Parallel and Subjective Time Scales --
Katsunari Shibata (Oita Univ.)
Abstract (in Japanese) (See Japanese page) 
(in English) As a general method for effective retrospective learning in uninterrupted time based on the concept of ``subjective time'', ``causality trace'' is introduced. A trace, which is assigned at each connection in each neuron, takes in the corresponding input signal according to the temporal change in the neuron's output. This enables to memorize only past important events, to hold them in its local memory, and to learn the past processes effectively. Through learning, the criteria of what is important is acqquired, and the division of roles in the time axis among neurons is promoted. From the viewpoint of time, there are parallel, non-uniform and subjective time scales in the neural network. The causality traces can be applied to value learning with a neural network, and also to the learning of recurrent neural networks though the way of application is a bit different. A new simulation result in a value-learning task shows its effectiveness and the division of roles in the time axis among neurons through learning.
Keyword (in Japanese) (See Japanese page) 
(in English) Causality Trace / Neural Network / Reinforcement Learning / Eligibility Trace / Subjective Time / Retrospective Learning / Recurrent Neural Network / Supervised Learning  
Reference Info. IEICE Tech. Rep., vol. 113, no. 500, NC2013-115, pp. 157-162, March 2014.
Paper # NC2013-115 
Date of Issue 2014-03-10 (NC) 
ISSN Print edition: ISSN 0913-5685  Online edition: ISSN 2432-6380

Conference Information
Committee NC MBE  
Conference Date 2014-03-17 - 2014-03-18 
Place (in Japanese) (See Japanese page) 
Place (in English) Tamagawa University 
Topics (in Japanese) (See Japanese page) 
Topics (in English)  
Paper Information
Registration To NC 
Conference Code 2014-03-NC-MBE 
Language Japanese 
Title (in Japanese) (See Japanese page) 
Sub Title (in Japanese) (See Japanese page) 
Title (in English) Causality Trace 
Sub Title (in English) Effective Retrospective Learning by Introducing Parallel and Subjective Time Scales 
Keyword(1) Causality Trace  
Keyword(2) Neural Network  
Keyword(3) Reinforcement Learning  
Keyword(4) Eligibility Trace  
Keyword(5) Subjective Time  
Keyword(6) Retrospective Learning  
Keyword(7) Recurrent Neural Network  
Keyword(8) Supervised Learning  
1st Author's Name Katsunari Shibata  
1st Author's Affiliation Oita University (Oita Univ.)
2nd Author's Name  
2nd Author's Affiliation ()
3rd Author's Name  
3rd Author's Affiliation ()
4th Author's Name  
4th Author's Affiliation ()
5th Author's Name  
5th Author's Affiliation ()
6th Author's Name  
6th Author's Affiliation ()
7th Author's Name  
7th Author's Affiliation ()
8th Author's Name  
8th Author's Affiliation ()
9th Author's Name  
9th Author's Affiliation ()
10th Author's Name  
10th Author's Affiliation ()
11th Author's Name  
11th Author's Affiliation ()
12th Author's Name  
12th Author's Affiliation ()
13th Author's Name  
13th Author's Affiliation ()
14th Author's Name  
14th Author's Affiliation ()
15th Author's Name  
15th Author's Affiliation ()
16th Author's Name  
16th Author's Affiliation ()
17th Author's Name  
17th Author's Affiliation ()
18th Author's Name  
18th Author's Affiliation ()
19th Author's Name  
19th Author's Affiliation ()
20th Author's Name  
20th Author's Affiliation ()
Speaker
Date Time 2014-03-18 16:00:00 
Presentation Time 20 
Registration for NC 
Paper # IEICE-NC2013-115 
Volume (vol) IEICE-113 
Number (no) no.500 
Page pp.157-162 
#Pages IEICE-6 
Date of Issue IEICE-NC-2014-03-10 


[Return to Top Page]

[Return to IEICE Web Page]


The Institute of Electronics, Information and Communication Engineers (IEICE), Japan