(Enter summary)
Abstract: In spite of the importance of representation
in learning, little progress has been made toward
understanding what makes representations
work. This paper describes a framework
for knowledge-level analysis of changes
in the representation of training examples in
concept learning. This a very fundamental
sort of representation change; such a change
alters the very space over which learning occurs,
and hence necessitates selection of a
new hypothesis space and (probably) a new
learning algorithm.... (Update)
Context of citations to this paper: More
.... would be to try to change the representation in order to find a representation where learning would be less complex (Subramanian 1989; Cohen 1990; Giordana and Saitta 1990) Using the linearity hypothesis, it is possible to introduce a boolean A molecule m17 One...
...necessary for EBL TS to work. This paper is an expanded version of [Cohen, 1990d] Section 3. 1 also draws on results presented in [Cohen, 1990a] 1.2 Overview Before presenting A EBL, we would like to clearly state the problem that it is intended to solve. It is not...
Cited by: More
ZooM: a nested Galois lattices-based system for.. - Pernelle, Rousset.. (2002)
(Correct)
Abductive Explanation-Based Learning: A Solution to the Multiple.. - Cohen (1994)
(Correct)
Solving multiple-instance and multiple-part learning problems with .. - Yann
(Correct)
Active bibliography (related documents): More All
0.8: Selective Reformulation of Examples in Concept Learning - Zucker, Ganascia (1994)
(Correct)
0.5: Evaluation and Selection of Biases in Machine Learning - desJardins, Gordon (1995)
(Correct)
0.2: The Generality Of Overgenerality - Cohen (1995)
(Correct)
Similar documents based on text: More All
0.1: Learning from Textbook Knowledge: A Case Study - Cohen (1990)
(Correct)
0.0: Learning Control Knowledge in Models of Expertise - Straatman (1995)
(Correct)
0.0: Applications of a Logic of Knowledge to Motion.. - Brafman, Latombe..
(Correct)
Related documents from co-citation: More All
4: Induction in abstraction spaces: A form of constructive induction (context) - Drastal, Czako et al. - 1989
3: Representational issues in machine learning (context) - Subramanian - 1989
3: Learning from textbook knowledge: A case study
- Cohen - 1990
BibTeX entry: (Update)
WilliamW. Cohen. An analysis of representation shift in concept learning. In Proceedings of the Seventh International Conference on Machine Learning, Austin, Texas, 1990. Morgan Kaufmann. http://citeseer.ist.psu.edu/article/cohen90analysis.html More
@inproceedings{ cohen90analysis,
author = "William W. Cohen",
title = "An Analysis of Representation Shift in Concept Learning",
booktitle = "Machine Learning",
pages = "104-112",
year = "1990",
url = "citeseer.ist.psu.edu/article/cohen90analysis.html" }
Citations (may not include all citations):
493
Communications of the ACM (context) - Valiant, of et al. - 1984
441
Queries and concept learning (context) - Angluin - 1988
412
Language identification in the limit (context) - Gold - 1967
371
Explanation-based generalization: A unifying view (context) - Mitchell, Keller et al. - 1986
317
Learning quickly when irrelevant attributes abound: A new li.. (context) - Littlestone - 1988
80
Shift of bias for inductive concept learning (context) - Utgoff - 1984
54
A study of explanation-based methods for inductive learning (context) - Flann, Dietterich - 1989
28
Artificial Intelligence (context) - Newell, level - 1982
24
Learning from textbook knowledge: A case study
- Cohen - 1990
22
Incremental version space merging: A general framework for c.. (context) - Hirsh - 1989
21
Induction in abstraction spaces: A form of constructive indu.. (context) - Drastal, Czako et al. - 1989
21
Toward a model of representation change (context) - Korf - 1981
20
Abductive explanation based learning: A solution to the mult..
- Cohen - 1989
12
Explanation-based generalization as resolution theorem provi.. (context) - KedarCabelli, McCarty - 1987
8
Generalizing number and learning from multiple examples in e..
- Cohen - 1988
6
An analysis of representation shift in concept learning
- Cohen - 1990
6
Representational issues in machine learning (context) - Subramanian - 1989
2
A mathematical framework for studying representations (context) - Holte, Zimmer - 1989
2
Evaluating bias during pac-learning (context) - Chrisman - 1989
1
Declarative bias for structural domains (context) - Russell, Grosof - 1989
1
On representations of problems of readings about actions (context) - Amarel - 1981
1
Compiling learning vocabulary from performance description (context) - Keller - 1989
1
EBL: An alternative view (context) - DeJong, Mooney - 1986
1
STRATA: Problem reformulation and abstract data types (context) - Lowry - 1988
1
Learning from being told and learning from examples: an expe.. (context) - Michalski, Chilausky - 1980
1
Reformulation from state space to reduction space (context) - Riddle - 1989
The graph only includes citing articles where the year of publication is known.
Documents on the same site (http://portal.research.bell-labs.com/orgs/ssr/people/wcohen/pubs.html): More
Text Categorization and Relational Learning - Cohen (1995)
(Correct)
Compiling Prior Knowledge Into an Explicit Bias - Cohen (1992)
(Correct)
Adaptive Mapping and Navigation by Teams of Simple Robots - Cohen (1996)
(Correct)
Online articles have much greater impact More about CiteSeer.IST Add search form to your site Submit documents Feedback
CiteSeer.IST - Copyright Penn State and NEC