Limits...
On Pixel-Wise Explanations for Non-Linear Classifier Decisions by Layer-Wise Relevance Propagation.

Bach S, Binder A, Montavon G, Klauschen F, Müller KR, Samek W - PLoS ONE (2015)

Bottom Line: Although machine learning methods are solving very successfully a plethora of tasks, they have in most cases the disadvantage of acting as a black box, not providing any information about what made them arrive at a particular decision.These pixel contributions can be visualized as heatmaps and are provided to a human expert who can intuitively not only verify the validity of the classification decision, but also focus further analysis on regions of potential interest.We evaluate our method for classifiers trained on PASCAL VOC 2009 images, synthetic image data containing geometric shapes, the MNIST handwritten digits data set and for the pre-trained ImageNet model available as part of the Caffe open source package.

View Article: PubMed Central - PubMed

Affiliation: Machine Learning Group, Fraunhofer Heinrich Hertz Institute, Berlin, Germany; Machine Learning Group, Technische Universität Berlin, Berlin, Germany.

ABSTRACT
Understanding and interpreting classification decisions of automated image classification systems is of high value in many applications, as it allows to verify the reasoning of the system and provides additional information to the human expert. Although machine learning methods are solving very successfully a plethora of tasks, they have in most cases the disadvantage of acting as a black box, not providing any information about what made them arrive at a particular decision. This work proposes a general solution to the problem of understanding classification decisions by pixel-wise decomposition of nonlinear classifiers. We introduce a methodology that allows to visualize the contributions of single pixels to predictions for kernel-based classifiers over Bag of Words features and for multilayered neural networks. These pixel contributions can be visualized as heatmaps and are provided to a human expert who can intuitively not only verify the validity of the classification decision, but also focus further analysis on regions of potential interest. We evaluate our method for classifiers trained on PASCAL VOC 2009 images, synthetic image data containing geometric shapes, the MNIST handwritten digits data set and for the pre-trained ImageNet model available as part of the Caffe open source package.

No MeSH data available.


Local and global predictions for input images are obtained by following a series of steps through the classification- and pixel-wise decomposition pipelines.Each step taken towards the final pixel-wise decomposition has a complementing analogue within the Bag of Words classification pipeline. The calculations used during the pixel-wise decomposition process make use of information extracted by those corresponding analogues. Airplane image in the graphic by Pixabay user tpsdave.
© Copyright Policy
Related In: Results  -  Collection

License
getmorefigures.php?uid=PMC4498753&req=5

pone.0130140.g004: Local and global predictions for input images are obtained by following a series of steps through the classification- and pixel-wise decomposition pipelines.Each step taken towards the final pixel-wise decomposition has a complementing analogue within the Bag of Words classification pipeline. The calculations used during the pixel-wise decomposition process make use of information extracted by those corresponding analogues. Airplane image in the graphic by Pixabay user tpsdave.

Mentions: The main contribution of this part is the formulation of a generic framework for retracing the origins of a decision made by the learned kernel-based classifier function for a BoW feature. This is achieved, in a broad sense as visualized in Fig 4, by following the construction of a BoW representation x of an image and the evaluation thereof by a classifier function in reverse direction. In this section we will derive a decomposition of a kernel-based classifier prediction into contributions of individual local features and finally single pixels. The proposed approach consists of three consecutive steps.


On Pixel-Wise Explanations for Non-Linear Classifier Decisions by Layer-Wise Relevance Propagation.

Bach S, Binder A, Montavon G, Klauschen F, Müller KR, Samek W - PLoS ONE (2015)

Local and global predictions for input images are obtained by following a series of steps through the classification- and pixel-wise decomposition pipelines.Each step taken towards the final pixel-wise decomposition has a complementing analogue within the Bag of Words classification pipeline. The calculations used during the pixel-wise decomposition process make use of information extracted by those corresponding analogues. Airplane image in the graphic by Pixabay user tpsdave.
© Copyright Policy
Related In: Results  -  Collection

License
Show All Figures
getmorefigures.php?uid=PMC4498753&req=5

pone.0130140.g004: Local and global predictions for input images are obtained by following a series of steps through the classification- and pixel-wise decomposition pipelines.Each step taken towards the final pixel-wise decomposition has a complementing analogue within the Bag of Words classification pipeline. The calculations used during the pixel-wise decomposition process make use of information extracted by those corresponding analogues. Airplane image in the graphic by Pixabay user tpsdave.
Mentions: The main contribution of this part is the formulation of a generic framework for retracing the origins of a decision made by the learned kernel-based classifier function for a BoW feature. This is achieved, in a broad sense as visualized in Fig 4, by following the construction of a BoW representation x of an image and the evaluation thereof by a classifier function in reverse direction. In this section we will derive a decomposition of a kernel-based classifier prediction into contributions of individual local features and finally single pixels. The proposed approach consists of three consecutive steps.

Bottom Line: Although machine learning methods are solving very successfully a plethora of tasks, they have in most cases the disadvantage of acting as a black box, not providing any information about what made them arrive at a particular decision.These pixel contributions can be visualized as heatmaps and are provided to a human expert who can intuitively not only verify the validity of the classification decision, but also focus further analysis on regions of potential interest.We evaluate our method for classifiers trained on PASCAL VOC 2009 images, synthetic image data containing geometric shapes, the MNIST handwritten digits data set and for the pre-trained ImageNet model available as part of the Caffe open source package.

View Article: PubMed Central - PubMed

Affiliation: Machine Learning Group, Fraunhofer Heinrich Hertz Institute, Berlin, Germany; Machine Learning Group, Technische Universität Berlin, Berlin, Germany.

ABSTRACT
Understanding and interpreting classification decisions of automated image classification systems is of high value in many applications, as it allows to verify the reasoning of the system and provides additional information to the human expert. Although machine learning methods are solving very successfully a plethora of tasks, they have in most cases the disadvantage of acting as a black box, not providing any information about what made them arrive at a particular decision. This work proposes a general solution to the problem of understanding classification decisions by pixel-wise decomposition of nonlinear classifiers. We introduce a methodology that allows to visualize the contributions of single pixels to predictions for kernel-based classifiers over Bag of Words features and for multilayered neural networks. These pixel contributions can be visualized as heatmaps and are provided to a human expert who can intuitively not only verify the validity of the classification decision, but also focus further analysis on regions of potential interest. We evaluate our method for classifiers trained on PASCAL VOC 2009 images, synthetic image data containing geometric shapes, the MNIST handwritten digits data set and for the pre-trained ImageNet model available as part of the Caffe open source package.

No MeSH data available.