SEQUENTIAL INSTANCE-BASED LEARNING FOR PLANNING IN THE CONTEXT OF AN IMPERFECT INFORMATION GAME

دانشگاه آزاد اسلامی اصفهان(خوراسگان)

منو

رکورد قبلی

رکورد بعدی

" SEQUENTIAL INSTANCE-BASED LEARNING FOR PLANNING IN THE CONTEXT OF AN IMPERFECT INFORMATION GAME "

Document Type	:	Latin Dissertation
Language of Document	:	English
Record Number	:	151430
Doc. No	:	ET23222
Main Entry	:	Jenngang Shih
Title Proper	:	SEQUENTIAL INSTANCE-BASED LEARNING FOR PLANNING IN THE CONTEXT OF AN IMPERFECT INFORMATION GAME
Note	:	This document is digital این مدرک بصورت الکترونیکی می باشد
Abstract	:	Finding sequential concepts, as in planning, is a complex task because of the exponentialsize of the search space. Empirical learning is an effective way to find sequential conceptsfrom observations. Sequential Instance-Based Learning (SIBL), which is presented here.is an empirical learning approach. modeled after instance-Based Learning (IBL), thatlearns sequential concepts, ordered sequences of state-action pairs to perform a synthesistask. SIBL is highly effective: it learns expert-level knowledge. SIBL demonstrates thefeasibility of using an empirical learning approach to discover sequential concepts. In ad-dition. this approach suggests a general framework that systematically extends empiricallearning to learning sequential concepts. In this dissertation. SLBL is tested on the domainof bridge.....-....,....-...,..tested for theQ1 PC1 bus cardBoth these projects mere sofixare des elopment efforts tonards contributing to dlfferentaspects of Roboucs and lZ1echatronics projects m the Controls and Roboucs Group..
Subject	:	Electericl tess
	:	برق
electronic file name	:	TL46458.pdf
Title and statement of responsibility and	:	SEQUENTIAL INSTANCE-BASED LEARNING FOR PLANNING IN THE CONTEXT OF AN IMPERFECT INFORMATION GAME [Thesis]