Cartoon Explanations of Image Classifiers

. We present CartoonX (Cartoon Explanation), a novel model-agnostic explanation method tailored towards image classifiers and based on the rate-distortion explanation (RDE) framework. Natural images are roughly piece-wise smooth signals—also called cartoon-like images—and tend to be sparse in the wavelet domain. CartoonX is the first explanation method to exploit this by requiring its explanations to be sparse in the wavelet domain, thus extracting the relevant piece-wise smooth part of an image instead of relevant pixel-sparse regions. We demonstrate that CartoonX can reveal novel valuable explanatory information, particularly for misclassifications. Moreover, we show that CartoonX achieves a lower distortion with fewer coefficients than state-of-the-art methods.

[1]  Giuseppe Caire,et al.  RadioUNet: Fast Radio Map Estimation With Convolutional Neural Networks , 2019, IEEE Transactions on Wireless Communications.

[2]  Gitta Kutyniok,et al.  In-Distribution Interpretability for Challenging Modalities , 2020, ArXiv.

[3]  Marissa Connor,et al.  Generative causal explanations of black-box classifiers , 2020, NeurIPS.

[4]  Arun Das,et al.  Opportunities and Challenges in Explainable Artificial Intelligence (XAI): A Survey , 2020, ArXiv.

[5]  Giuseppe Caire,et al.  Pathloss Prediction using Deep Learning with Applications to Cellular Optimization and Efficient D2D Link Scheduling , 2020, ICASSP 2020 - 2020 IEEE International Conference on Acoustics, Speech and Signal Processing (ICASSP).

[6]  Andrea Vedaldi,et al.  Understanding Deep Networks via Extremal Perturbations and Smooth Masks , 2019, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

[7]  Gitta Kutyniok,et al.  A Rate-Distortion Framework for Explaining Neural Network Decisions , 2019, ArXiv.

[8]  Quoc V. Le,et al.  Searching for MobileNetV3 , 2019, 2019 IEEE/CVF International Conference on Computer Vision (ICCV).

[9]  David Duvenaud,et al.  Explaining Image Classifiers by Counterfactual Generation , 2018, ICLR.

[10]  Mukund Sundararajan,et al.  How Important Is a Neuron? , 2018, ICLR.

[11]  Kate Saenko,et al.  RISE: Randomized Input Sampling for Explanation of Black-box Models , 2018, BMVC.

[12]  Martin Wattenberg,et al.  Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV) , 2017, ICML.

[13]  Martin Wattenberg,et al.  SmoothGrad: removing noise by adding noise , 2017, ArXiv.

[14]  Scott Lundberg,et al.  A Unified Approach to Interpreting Model Predictions , 2017, NIPS.

[15]  Yarin Gal,et al.  Real Time Image Saliency for Black Box Classifiers , 2017, NIPS.

[16]  Andrea Vedaldi,et al.  Interpretable Explanations of Black Boxes by Meaningful Perturbation , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[17]  Avanti Shrikumar,et al.  Learning Important Features Through Propagating Activation Differences , 2017, ICML.

[18]  Karen Simonyan,et al.  Neural Audio Synthesis of Musical Notes with WaveNet Autoencoders , 2017, ICML.

[19]  Ankur Taly,et al.  Axiomatic Attribution for Deep Networks , 2017, ICML.

[20]  Abhishek Das,et al.  Grad-CAM: Visual Explanations from Deep Networks via Gradient-Based Localization , 2016, 2017 IEEE International Conference on Computer Vision (ICCV).

[21]  Thomas Brox,et al.  Synthesizing the preferred inputs for neurons in neural networks via deep generator networks , 2016, NIPS.

[22]  Carlos Guestrin,et al.  "Why Should I Trust You?": Explaining the Predictions of Any Classifier , 2016, ArXiv.

[23]  Sergey Ioffe,et al.  Rethinking the Inception Architecture for Computer Vision , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[24]  Alexander Binder,et al.  On Pixel-Wise Explanations for Non-Linear Classifier Decisions by Layer-Wise Relevance Propagation , 2015, PloS one.

[25]  Jimmy Ba,et al.  Adam: A Method for Stochastic Optimization , 2014, ICLR.

[26]  Thomas Brox,et al.  Striving for Simplicity: The All Convolutional Net , 2014, ICLR.

[27]  Andrew Zisserman,et al.  Very Deep Convolutional Networks for Large-Scale Image Recognition , 2014, ICLR.

[28]  Andrew Zisserman,et al.  Deep Inside Convolutional Networks: Visualising Image Classification Models and Saliency Maps , 2013, ICLR.

[29]  Geoffrey E. Hinton,et al.  ImageNet classification with deep convolutional neural networks , 2012, Commun. ACM.

[30]  Wang-Q Lim,et al.  Compactly supported shearlets are optimally sparse , 2010, J. Approx. Theory.

[31]  Stphane Mallat,et al.  A Wavelet Tour of Signal Processing, Third Edition: The Sparse Way , 2008 .

[32]  Justin K. Romberg,et al.  Wavelet-domain approximation and compression of piecewise smooth images , 2006, IEEE Transactions on Image Processing.

[33]  R. DeVore,et al.  Nonlinear approximation , 1998, Acta Numerica.

[34]  S. Mallat.  A wavelet tour of signal processing , 1998 .