Abstract Understanding behavior of neural networks is necessary in order to better analyze and diagnose them. Quantitative metrics such as classification accuracy and F1 score just give us numbers indicating how good is the classifier in our problem. They do not tell us how a neural network achieves this result. Visualization is a set of techniques that are commonly used for understanding structure of high-dimensional vectors. In this chapter, we briefly reviewed data-driven techniques for visualization and showed that how to apply them on neural networks. Then, we focused on techniques that visualize neural networks by minimizing an objective function. Among them, we explained three different methods. In the first method, we defined a loss function and found an image that maximizes the classification score of a particular class. In order to generate more interpretable images, the objective function was regularized using norm of the image. In the second method, gradient of a particular neuron was computed with respect to the input image and it is illustrated by computing its magnitude. The third method formulated the visualizing problem as an image reconstruction problem. To be more specific, we explained a method that tries to find an image in which the representation of this image is very close to the representation of the original image. This technique usually tells us what information is usually discarded by a particular layer.
Visualizing Neural Networks
Guide to Convolutional Neural Networks ; 247-258
2017-01-01
12 pages
Article/Chapter (Book)
Electronic Resource
English
Neural Network , Feature Space , Input Image , Local Binary Pattern , Classification Score Computer Science , Pattern Recognition , Information Systems Applications (incl. Internet) , Computer Systems Organization and Communication Networks , Signal, Image and Speech Processing , Language Translation and Linguistics , Automotive Engineering
Visualizing Convolutional Neural Networks with Virtual Reality
German Aerospace Center (DLR) | 2019
|Visualizing Deep Convolutional Neural Networks Using Natural Pre-images
British Library Online Contents | 2016
|Visualizing Evolving Networks: Minimum Spanning Trees versus Pathfinder Networks
British Library Conference Proceedings | 2003
|British Library Online Contents | 1994