N-learners Problem: Fusion of Concepts

N. S.V. Rao; E. M. Oblow; C. W. Glover; G. E. Liepins

doi:10.1109/IROS.1992.594563

N-learners Problem: Fusion of Concepts

N. S.V. Rao, E. M. Oblow, C. W. Glover, G. E. Liepins

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

1 Scopus citations

Abstract

Given N learners each capable of learning concepts (subsets) of a domain set X in the sense of Valiant [19], we are interested in obtaining a composite system constituted by the fuser and the individual learners. We consider two cases: open and closed fusion. In open fusion the fuser is given the sample and the hypotheses of the individual learners; we show that the fusion rule can be obtained by formulating this problem as another learning problem. For the case all individual learners are trained with the same sample, we show sufficiency conditions that ensure the composite system to be better than the best of the individual. Second, in closed fusion the fuser does not have an access to either the training sample or the hypotheses of the individual learners. By suitably designing a linear threshold function of the outputs of individual learners, we show that the composite system can be made better than the best of the learners.

Original language	English
Title of host publication	IROS 1992 - Proceedings of the 1992 IEEE/RSJ International Conference on Intelligent Robots and Systems
Subtitle of host publication	Sensor-Based Robotics and Opportunties for its Industrial Applications
Publisher	Institute of Electrical and Electronics Engineers Inc.
Pages	1372-1379
Number of pages	8
ISBN (Electronic)	0780307372
DOIs	https://doi.org/10.1109/IROS.1992.594563
State	Published - 1992
Externally published	Yes
Event	1992 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 1992 - Raleigh, United States Duration: Jul 7 1992 → Jul 10 1992

Publication series

Name	IEEE International Conference on Intelligent Robots and Systems
Volume	2
ISSN (Print)	2153-0858
ISSN (Electronic)	2153-0866

Conference

Conference	1992 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 1992
Country/Territory	United States
City	Raleigh
Period	07/7/92 → 07/10/92

Funding

The authors gratefully acknowledge the continuing financial support of this learning research by Oscar Manley of the Basic Energy Sciences Program in the Department of Energy and Teresa McMullen in the Intelligent Systems Program of the Office of Naval Research in the Department of Defense. Also, the first author is funded by National Science Foundation under grant #IRI-9108610, Oak Ridge National Laboratory operated by Martin Marietta under contracts #19X-SE043V and #SOXSJ433V, Old Dominion University Summer Faculty Fellowship for 1991 and Virginia's Center for Innovative Technology under contract # INF-90-015.

Access to Document

10.1109/IROS.1992.594563

Cite this

Rao, N. S. V., Oblow, E. M., Glover, C. W., & Liepins, G. E. (1992). N-learners Problem: Fusion of Concepts. In IROS 1992 - Proceedings of the 1992 IEEE/RSJ International Conference on Intelligent Robots and Systems: Sensor-Based Robotics and Opportunties for its Industrial Applications (pp. 1372-1379). Article 594563 (IEEE International Conference on Intelligent Robots and Systems; Vol. 2). Institute of Electrical and Electronics Engineers Inc.. https://doi.org/10.1109/IROS.1992.594563

Rao, N. S.V. ; Oblow, E. M. ; Glover, C. W. et al. / N-learners Problem : Fusion of Concepts. IROS 1992 - Proceedings of the 1992 IEEE/RSJ International Conference on Intelligent Robots and Systems: Sensor-Based Robotics and Opportunties for its Industrial Applications. Institute of Electrical and Electronics Engineers Inc., 1992. pp. 1372-1379 (IEEE International Conference on Intelligent Robots and Systems).

@inproceedings{3c05982a3bbb4026844dcd53da92c367,

title = "N-learners Problem: Fusion of Concepts",

abstract = "Given N learners each capable of learning concepts (subsets) of a domain set X in the sense of Valiant [19], we are interested in obtaining a composite system constituted by the fuser and the individual learners. We consider two cases: open and closed fusion. In open fusion the fuser is given the sample and the hypotheses of the individual learners; we show that the fusion rule can be obtained by formulating this problem as another learning problem. For the case all individual learners are trained with the same sample, we show sufficiency conditions that ensure the composite system to be better than the best of the individual. Second, in closed fusion the fuser does not have an access to either the training sample or the hypotheses of the individual learners. By suitably designing a linear threshold function of the outputs of individual learners, we show that the composite system can be made better than the best of the learners.",

author = "Rao, {N. S.V.} and Oblow, {E. M.} and Glover, {C. W.} and Liepins, {G. E.}",

note = "Publisher Copyright: {\textcopyright} 1992 IEEE.; 1992 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 1992 ; Conference date: 07-07-1992 Through 10-07-1992",

year = "1992",

doi = "10.1109/IROS.1992.594563",

language = "English",

series = "IEEE International Conference on Intelligent Robots and Systems",

publisher = "Institute of Electrical and Electronics Engineers Inc.",

pages = "1372--1379",

booktitle = "IROS 1992 - Proceedings of the 1992 IEEE/RSJ International Conference on Intelligent Robots and Systems",

}

Rao, NSV, Oblow, EM, Glover, CW & Liepins, GE 1992, N-learners Problem: Fusion of Concepts. in IROS 1992 - Proceedings of the 1992 IEEE/RSJ International Conference on Intelligent Robots and Systems: Sensor-Based Robotics and Opportunties for its Industrial Applications., 594563, IEEE International Conference on Intelligent Robots and Systems, vol. 2, Institute of Electrical and Electronics Engineers Inc., pp. 1372-1379, 1992 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 1992, Raleigh, United States, 07/7/92. https://doi.org/10.1109/IROS.1992.594563

N-learners Problem: Fusion of Concepts. / Rao, N. S.V.; Oblow, E. M.; Glover, C. W. et al.
IROS 1992 - Proceedings of the 1992 IEEE/RSJ International Conference on Intelligent Robots and Systems: Sensor-Based Robotics and Opportunties for its Industrial Applications. Institute of Electrical and Electronics Engineers Inc., 1992. p. 1372-1379 594563 (IEEE International Conference on Intelligent Robots and Systems; Vol. 2).

Research output: Chapter in Book/Report/Conference proceeding › Conference contribution › peer-review

TY - GEN

T1 - N-learners Problem

T2 - 1992 IEEE/RSJ International Conference on Intelligent Robots and Systems, IROS 1992

AU - Rao, N. S.V.

AU - Oblow, E. M.

AU - Glover, C. W.

AU - Liepins, G. E.

PY - 1992

Y1 - 1992

N2 - Given N learners each capable of learning concepts (subsets) of a domain set X in the sense of Valiant [19], we are interested in obtaining a composite system constituted by the fuser and the individual learners. We consider two cases: open and closed fusion. In open fusion the fuser is given the sample and the hypotheses of the individual learners; we show that the fusion rule can be obtained by formulating this problem as another learning problem. For the case all individual learners are trained with the same sample, we show sufficiency conditions that ensure the composite system to be better than the best of the individual. Second, in closed fusion the fuser does not have an access to either the training sample or the hypotheses of the individual learners. By suitably designing a linear threshold function of the outputs of individual learners, we show that the composite system can be made better than the best of the learners.

AB - Given N learners each capable of learning concepts (subsets) of a domain set X in the sense of Valiant [19], we are interested in obtaining a composite system constituted by the fuser and the individual learners. We consider two cases: open and closed fusion. In open fusion the fuser is given the sample and the hypotheses of the individual learners; we show that the fusion rule can be obtained by formulating this problem as another learning problem. For the case all individual learners are trained with the same sample, we show sufficiency conditions that ensure the composite system to be better than the best of the individual. Second, in closed fusion the fuser does not have an access to either the training sample or the hypotheses of the individual learners. By suitably designing a linear threshold function of the outputs of individual learners, we show that the composite system can be made better than the best of the learners.

UR - http://www.scopus.com/inward/record.url?scp=85051413619&partnerID=8YFLogxK

U2 - 10.1109/IROS.1992.594563

DO - 10.1109/IROS.1992.594563

M3 - Conference contribution

AN - SCOPUS:85051413619

T3 - IEEE International Conference on Intelligent Robots and Systems

SP - 1372

EP - 1379

BT - IROS 1992 - Proceedings of the 1992 IEEE/RSJ International Conference on Intelligent Robots and Systems

PB - Institute of Electrical and Electronics Engineers Inc.

Y2 - 7 July 1992 through 10 July 1992

ER -

Rao NSV, Oblow EM, Glover CW, Liepins GE. N-learners Problem: Fusion of Concepts. In IROS 1992 - Proceedings of the 1992 IEEE/RSJ International Conference on Intelligent Robots and Systems: Sensor-Based Robotics and Opportunties for its Industrial Applications. Institute of Electrical and Electronics Engineers Inc. 1992. p. 1372-1379. 594563. (IEEE International Conference on Intelligent Robots and Systems). doi: 10.1109/IROS.1992.594563

N-learners Problem: Fusion of Concepts

Abstract

Publication series

Conference

Funding

Access to Document

Other files and links

Fingerprint

Cite this