TY - GEN
T1 - Analyzing Classifiers
T2 - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016
AU - Lapuschkin, Sebastian
AU - Binder, Alexander
AU - Montavon, Gregoire
AU - Muller, Klaus
AU - Samek, Wojciech
PY - 2016/12/9
Y1 - 2016/12/9
N2 - Fisher vector (FV) classifiers and Deep Neural Networks (DNNs) are popular and successful algorithms for solving image classification problems. However, both are generally considered 'black box' predictors as the non-linear transformations involved have so far prevented transparent and interpretable reasoning. Recently, a principled technique, Layer-wise Relevance Propagation (LRP), has been developed in order to better comprehend the inherent structured reasoning of complex nonlinear classification models such as Bag of Feature models or DNNs. In this paper we (1) extend the LRP framework also for Fisher vector classifiers and then use it as analysis tool to (2) quantify the importance of context for classification, (3) qualitatively compare DNNs against FV classifiers in terms of important image regions and (4) detect potential flaws and biases in data. All experiments are performed on the PASCAL VOC 2007 and ILSVRC 2012 data sets.
AB - Fisher vector (FV) classifiers and Deep Neural Networks (DNNs) are popular and successful algorithms for solving image classification problems. However, both are generally considered 'black box' predictors as the non-linear transformations involved have so far prevented transparent and interpretable reasoning. Recently, a principled technique, Layer-wise Relevance Propagation (LRP), has been developed in order to better comprehend the inherent structured reasoning of complex nonlinear classification models such as Bag of Feature models or DNNs. In this paper we (1) extend the LRP framework also for Fisher vector classifiers and then use it as analysis tool to (2) quantify the importance of context for classification, (3) qualitatively compare DNNs against FV classifiers in terms of important image regions and (4) detect potential flaws and biases in data. All experiments are performed on the PASCAL VOC 2007 and ILSVRC 2012 data sets.
UR - http://www.scopus.com/inward/record.url?scp=84986268738&partnerID=8YFLogxK
UR - http://www.scopus.com/inward/citedby.url?scp=84986268738&partnerID=8YFLogxK
U2 - 10.1109/CVPR.2016.318
DO - 10.1109/CVPR.2016.318
M3 - Conference contribution
AN - SCOPUS:84986268738
T3 - Proceedings of the IEEE Computer Society Conference on Computer Vision and Pattern Recognition
SP - 2912
EP - 2920
BT - Proceedings - 29th IEEE Conference on Computer Vision and Pattern Recognition, CVPR 2016
PB - IEEE Computer Society
Y2 - 26 June 2016 through 1 July 2016
ER -