(Enter summary)
Abstract: We consider the question "How should one act when the only goal is to learn as much as possible?" Building
on the theoretical results of Fedorov [1972] and MacKay [1992], we apply techniques from Optimal
Experiment Design (OED) to guide the query/action selection of a neural network learner. We demonstrate
that these techniques allow the learner to minimize its generalization error by exploring its domain
efficiently and completely. We conclude that, while not a panacea, OED-based query/action... (Update)
Cited by: More
Active Learning for Multi-Class Logistic Regression - Schein, Ungar (2005)
(Correct)
Active Learning for Logistic Regression - Schein (2005)
(Correct)
Pattern Recognition as a Deterministic Problem: An Approach .. - Cervellera, Muselli
(Correct)
Similar documents (at the sentence level):
5.9%: Neural Network Exploration Using Optimal Experiment Design - David Cohn (1994)
(Correct)
Active bibliography (related documents): More All
0.5: Argument Raising Meets Adjuncts-as-Dependents and Traceless.. - De Kuthy, Meurers
(Correct)
0.4: Active Learning with Statistical Models - Cohn, Ghahramani, al. (1995)
(Correct)
0.3: Minimizing Statistical Bias with Queries - Cohn (1995)
(Correct)
Similar documents based on text: More All
0.1: Incremental Active Learning in. . . - Sugiyama, al.
(Correct)
0.1: Incremental Active Learning for Optimal Generalization - Sugiyama, Ogawa (2000)
(Correct)
Related documents from co-citation: More All
43: Information-based objective functions for active data selection
- MacKay
33: Theory of Optimal Experiments (context) - Fedorov - 1972
20: Active learning in multilayer perceptrons
- Fukumizu - 1996
BibTeX entry: (Update)
Cohn, DA (1994). Neural network exploration using optimal experiment design. In JD Cowan, G Tesauro & J Allspector, editors, Advances in Neural Information Processing Systems, 6. San Mateo, CA: Morgan Kaufmann, 679-686. http://citeseer.ist.psu.edu/article/cohn94neural.html More
@inproceedings{ cohn94neural,
author = "David A. Cohn",
title = "Neural Network Exploration Using Optimal Experiment Design",
booktitle = "Advances in Neural Information Processing Systems",
volume = "6",
publisher = "Morgan Kaufmann Publishers, Inc.",
editor = "Jack D. Cowan and Gerald Tesauro and Joshua Alspector",
pages = "679--686",
year = "1994",
url = "citeseer.ist.psu.edu/article/cohn94neural.html" }
Citations (may not include all citations):
465
Learnability and the VapnikChervonenkis dimension (context) - Blumer, Ehrenfeucht et al. - 1989
413
Adaptive mixtures of local experts (context) - Jacobs, Jordan et al. - 1991
240
Advances in Neural Information Processing Systems (context) - Hanson - 1992
240
Advances in Neural Information Processing Systems (context) - Touretzky, Advances et al. - 1994
205
Mixture Models: Inference and Applications to Clustering (context) - McLachlan, Basford - 1988
203
What size net gives valid generalization (context) - Baum, Haussler - 1989
135
Introduction to robotics (context) - Craig - 1989 ACM
116
Forward models: Supervised learning with a distal teacher
- Jordan, Rumelhart - 1992 DBLP
105
Information-based objective functions for active data select..
- MacKay - 1992 ACM
76
Supervised learning from incomplete data via an EM approach
- Ghahramani, Jordan - 1994 DBLP
48
Soft Competitive Adaptation: Neural Network Learning Algorit.. (context) - Nowlan - 1991
46
Selecting concise training sets from clean data (context) - Plutowski, White - 1993
39
Optimum experimental designs (context) - Atkinson - 1992
39
Training connectionist networks with queries and selective s.. (context) - Cohn, Atlas et al. - 1990 ACM DBLP
37
Robot Juggling: An Implementation of Memory-based Learning (context) - Schaal, Atkeson - 1994
36
Fast Exact Multiplication by the Hessian
- Pearlmutter - 1994 ACM
29
A note on the number of queries needed to identify regular l.. (context) - Angluin - 1982 DBLP
20
Neural model of adaptive hand-eye coordination for single po.. (context) - Kuperstein - 1988
19
the sample complexity of pac-learning using random and chose..
- Eisenberg, Rivest - 1990
18
Connectionist robot motion planning: a neurally-inspired app.. (context) - Mel - 1992
18
Constructing hidden units using examples and queries (context) - Baum, Lang - 1991 ACM DBLP
13
Generalizing the pac model for neural nets and other learnin.. (context) - Haussler - 1992
12
Reinforcement driven information acquisition in nondetermini..
- Schmidhuber, Storck - 1993
5
Query construction (context) - Sollich - 1994
4
Implementing Inner Drive by Competence Reflection (context) - Linden, Weber - 1993
4
On finding exciting trajectories for identification experime.. (context) - Armstrong - 1989
3
Bootstrap methods in neural network time series prediction (context) - Connor - 1993
3
Evolutionary operation (context) - Box, Draper - 1969
2
Doctoral dissertation (context) - New, Choueiki - 1994
1
Elements of Statistical Computing (context) - LEMS-, Engineering et al. - 1988 ACM
1
Active exploration in dynamic environments (context) - Hall, Thrun et al. - 1992 DBLP
1
ACM 3rd Annual Workshop on Computational Learning Theory (context) - Fulk, Case - 1972
1
Regression by local fitting (context) - Industrial, Engineering et al. - 1988
1
Practical reliable recognition of 2D and 3D objects using im.. (context) - Subrahmonia, Cooper et al. - 1992
The graph only includes citing articles where the year of publication is known.
Documents on the same site (http://www.neurop2.ruhr-uni-bochum.de/MIT_lit.html): More
MIT SchMUSE: Class-Based Remote Delegation in a.. - Blair, Cohen.. (1993)
(Correct)
The Logical Problem of Language Change - Niyogi (1995)
(Correct)
Aeroacoustics on Non-Dedicated Workstations - Skordos (1995)
(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