On concept space and hypothesis space in case-based learning algorithms
作者： A D Griffiths , D G Bridge
关键词:
摘要: In order to learn more about the behaviour of case-based reasoners as learning systems, we formalise a simple learner PAC algorithm. We show that representation ĄCB, σC is rich enough express any boolean function. define family algorithms which use single, fixed similarity measure and give necessary sufficient conditions for consistency these in terms chosen measure. Finally, consider way algorithms, when trained on target concepts from restricted concept space, often output hypotheses are outside space. A case study investigates this relationship between space hypothesis concludes algorithm studied less than optimal chosen, small,
springer.com
core.ac.uk
sci-hub.st 参考文章(13)
Christoph Globig, Stefan Wess, Symbolic Learning and Nearest-Neighbor Classification Springer, Berlin, Heidelberg. pp. 17- 27 ,(1994) , 10.1007/978-3-642-46808-7_2
Stefan Wess, Christoph Globig, Case-Based and Symbolic Classification - A Case Study EWCBR '93 Selected papers from the First European Workshop on Topics in Case-Based Reasoning. pp. 77- 91 ,(1993) , 10.1007/3-540-58330-0_78
David Haussler, Probably approximately correct learning national conference on artificial intelligence. pp. 1101- 1108 ,(1990)
Norman Biggs, Martin Anthony, Computational learning theory: an introduction Cambridge University Press. ,(1992)
Klaus P. Jantke, Steffen Lange, Case-Based Representation and Learning of Pattern Languages algorithmic learning theory. ,vol. 137, pp. 25- 51 ,(1993) , 10.1016/0304-3975(95)91134-C
David W. Aha, Marc K. Albert, Analyses of instance-based learning algorithms national conference on artificial intelligence. pp. 553- 558 ,(1991)
PETER TURNEY, Theoretical analyses of cross-validation error and voting in instance-based learning Journal of Experimental and Theoretical Artificial Intelligence. ,vol. 6, pp. 331- 360 ,(1994) , 10.1080/09528139408953793
David Haussler, Quantifying inductive bias: AI learning algorithms and Valiant's learning framework Artificial Intelligence. ,vol. 36, pp. 177- 221 ,(1988) , 10.1016/0004-3702(88)90002-1
Scott Cost, Steven Salzberg, A Weighted Nearest Neighbor Algorithm for Learning with Symbolic Features Machine Learning. ,vol. 10, pp. 57- 78 ,(1993) , 10.1023/A:1022664626993
William H.E. Day, Daniel P. Faith, A model in partial orders for comparing objects by dualistic measures Bellman Prize in Mathematical Biosciences. ,vol. 78, pp. 179- 192 ,(1986) , 10.1016/0025-5564(86)90123-9