USING REINFORCEMENT LEARNING FOR SIMILARITY ASSESSMENT IN CASE-BASED SYSTEMS

منو

" USING REINFORCEMENT LEARNING FOR SIMILARITY ASSESSMENT IN CASE-BASED SYSTEMS "

Document Type	:	Latin Dissertation
Language of Document	:	English
Record Number	:	149595
Doc. No	:	ET21387
Main Entry	:	Patrick Paulson
Title Proper	:	USING REINFORCEMENT LEARNING FOR SIMILARITY ASSESSMENT IN CASE-BASED SYSTEMS
Note	:	This document is digital این مدرک بصورت الکترونیکی می باشد
Abstract	:	This research adds reinforcement learning to the similarity assessmentcomponent of case-based reasoning systems (CBR). The argument is thatreinforcement learning allows case-based reasoning to be more easily used inknowledge-poor domains.The system uses a case base composed of cases that were previouslyencountered. Each case suggests a solution to be tried in a particular context. Inan application, the system is presented with the context for a new problemsituation. The system finds a similar context within its case base and returns thesuggested solution. The user then provides the system with a reinforcement signalwhich indicates how well the solution performs in the new context. Given thisfeedback, the system modifies how.
Subject	:	Electericl tess
	:	برق
electronic file name	:	TL44546.pdf
Title and statement of responsibility and	:	USING REINFORCEMENT LEARNING FOR SIMILARITY ASSESSMENT IN CASE-BASED SYSTEMS [Thesis]