Visualization of neural networks using saliency maps

Niels J.S. Mørch; Ulrik Kjems; Lars Kai Hansen; C. Svarer; I. Law; B. Lautrup; S. Strother; K. Rehm

doi:10.1109/ICNN.1995.488997

Visualization of neural networks using saliency maps

Niels J.S. Mørch, Ulrik Kjems, Lars Kai Hansen, C. Svarer, I. Law, B. Lautrup, S. Strother, K. Rehm

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

1123 Downloads (Pure)

Abstract

The saliency map is proposed as a new method for understanding and visualizing the nonlinearities embedded in feedforward neural networks, with emphasis on the ill-posed case, where the dimensionality of the input-field by far exceeds the number of examples. Several levels of approximations are discussed. The saliency maps are applied to medical imaging (PET-scans) for identification of paradigm-relevant regions in the human brain

Original language	English
Title of host publication	Neural Networks, 1995. Proceedings., IEEE International Conference on
Volume	4
Publisher	IEEE
Publication date	1995
Pages	2085-2090
ISBN (Print)	0-7803-2768-3
DOIs	https://doi.org/10.1109/ICNN.1995.488997
Publication status	Published - 1995
Event	1995 IEEE International Conference on Neural Networks - Perth, WA, United States Duration: 27 Nov 1995 → 1 Dec 1995 http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=3505

Conference

Conference	1995 IEEE International Conference on Neural Networks
Country/Territory	United States
City	Perth, WA
Period	27/11/1995 → 01/12/1995
Internet address	http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=3505

Bibliographical note

Copyright: 1995 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE

Access to Document

10.1109/ICNN.1995.488997

MørchFinal published version, 563 KB

Cite this

@inproceedings{a1bfb8e083464715a6bf2cff767b6062,

title = "Visualization of neural networks using saliency maps",

abstract = "The saliency map is proposed as a new method for understanding and visualizing the nonlinearities embedded in feedforward neural networks, with emphasis on the ill-posed case, where the dimensionality of the input-field by far exceeds the number of examples. Several levels of approximations are discussed. The saliency maps are applied to medical imaging (PET-scans) for identification of paradigm-relevant regions in the human brain",

author = "M{\o}rch, {Niels J.S.} and Ulrik Kjems and Hansen, {Lars Kai} and C. Svarer and I. Law and B. Lautrup and S. Strother and K. Rehm",

note = "Copyright: 1995 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE; 1995 IEEE International Conference on Neural Networks ; Conference date: 27-11-1995 Through 01-12-1995",

year = "1995",

doi = "10.1109/ICNN.1995.488997",

language = "English",

isbn = "0-7803-2768-3",

volume = "4",

pages = "2085--2090",

booktitle = "Neural Networks, 1995. Proceedings., IEEE International Conference on",

publisher = "IEEE",

address = "United States",

url = "http://ieeexplore.ieee.org/xpl/mostRecentIssue.jsp?punumber=3505",

}

Mørch, NJS, Kjems, U, Hansen, LK, Svarer, C, Law, I, Lautrup, B, Strother, S & Rehm, K 1995, Visualization of neural networks using saliency maps. in Neural Networks, 1995. Proceedings., IEEE International Conference on. vol. 4, IEEE, pp. 2085-2090, 1995 IEEE International Conference on Neural Networks, Perth, WA, United States, 27/11/1995. https://doi.org/10.1109/ICNN.1995.488997

Visualization of neural networks using saliency maps. / Mørch, Niels J.S.; Kjems, Ulrik; Hansen, Lars Kai; Svarer, C.; Law, I.; Lautrup, B.; Strother, S.; Rehm, K.

Neural Networks, 1995. Proceedings., IEEE International Conference on. Vol. 4 IEEE, 1995. p. 2085-2090.

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

TY - GEN

T1 - Visualization of neural networks using saliency maps

AU - Mørch, Niels J.S.

AU - Kjems, Ulrik

AU - Hansen, Lars Kai

AU - Svarer, C.

AU - Law, I.

AU - Lautrup, B.

AU - Strother, S.

AU - Rehm, K.

N1 - Copyright: 1995 IEEE. Personal use of this material is permitted. However, permission to reprint/republish this material for advertising or promotional purposes or for creating new collective works for resale or redistribution to servers or lists, or to reuse any copyrighted component of this work in other works must be obtained from the IEEE

PY - 1995

Y1 - 1995

N2 - The saliency map is proposed as a new method for understanding and visualizing the nonlinearities embedded in feedforward neural networks, with emphasis on the ill-posed case, where the dimensionality of the input-field by far exceeds the number of examples. Several levels of approximations are discussed. The saliency maps are applied to medical imaging (PET-scans) for identification of paradigm-relevant regions in the human brain

AB - The saliency map is proposed as a new method for understanding and visualizing the nonlinearities embedded in feedforward neural networks, with emphasis on the ill-posed case, where the dimensionality of the input-field by far exceeds the number of examples. Several levels of approximations are discussed. The saliency maps are applied to medical imaging (PET-scans) for identification of paradigm-relevant regions in the human brain

U2 - 10.1109/ICNN.1995.488997

DO - 10.1109/ICNN.1995.488997

M3 - Article in proceedings

SN - 0-7803-2768-3

VL - 4

SP - 2085

EP - 2090

BT - Neural Networks, 1995. Proceedings., IEEE International Conference on

PB - IEEE

T2 - 1995 IEEE International Conference on Neural Networks

Y2 - 27 November 1995 through 1 December 1995

ER -

Visualization of neural networks using saliency maps

Abstract

Conference

Bibliographical note

Access to Document

Fingerprint

Cite this