Presentation	1998/1/22 Learning Action Sequences Using Active Control of Multiresolutional Vision. Satoru MORITA,
PDF Download Page	PDF download Page Link
Abstract(in Japanese)	(See Japanese page)
Abstract(in English)	We propose the learning method of the action sequences using the image map gotten by the control of the multiresolutional vision. The recurrent neural net structure have the multiresolutional channel, the view, and the action. In general, the learning of the neural net have the pattern which can not be learned. The image map is generated by the control of the multiresolutional vision using Genetic Algorithm. The robot can acquire the action sequences by the stable learning of the recurrent neural net using the image map. It is found that the action sequences is acquired by the pan-tilt camera.
Keyword(in Japanese)	(See Japanese page)
Keyword(in English)	Multiresolusional Vision / Recurrent Neural Net / Genetic Algorithm
Paper #	PRMU97-200
Date of Issue

Conference Information
Committee	PRMU
Conference Date	1998/1/22(1days)
Place (in Japanese)	(See Japanese page)
Place (in English)
Topics (in Japanese)	(See Japanese page)
Topics (in English)
Chair
Vice Chair
Secretary
Assistant

Paper Information
Registration To	Pattern Recognition and Media Understanding (PRMU)
Language	JPN
Title (in Japanese)	(See Japanese page)
Sub Title (in Japanese)	(See Japanese page)
Title (in English)	Learning Action Sequences Using Active Control of Multiresolutional Vision.
Sub Title (in English)
Keyword(1)	Multiresolusional Vision
Keyword(2)	Recurrent Neural Net
Keyword(3)	Genetic Algorithm
1st Author's Name	Satoru MORITA
1st Author's Affiliation	Faculty of Engineering, Yamaguchi University()
Date	1998/1/22
Paper #	PRMU97-200
Volume (vol)	vol.97
Number (no)	500
Page	pp.pp.-
#Pages	7
Date of Issue