iNNvestigate neural networks!

Maximilian Alber
Maximilian Alber
Sebastian Lapuschkin
Sebastian Lapuschkin
Philipp Seegerer
Philipp Seegerer
Miriam Hägele
Miriam Hägele

Journal of Machine Learning Research, Volume abs/1808.04260, 2019.

Cited by: 59|Bibtex|Views111
EI
Other Links: dblp.uni-trier.de|arxiv.org
Weibo:
To demonstrate the versatility of iNNvestigate, we provide an analysis of image classifications for variety of state-of-the-art neural network architectures

Abstract:

In recent years, deep neural networks have revolutionized many application domains of machine learning and are key components of many critical decision or predictive processes. Therefore, it is crucial that domain specialists can understand and analyze actions and pre- dictions, even of the most complex neural network architectures. Des...More

Code:

Data:

0
Introduction
  • In recent years deep neural networks have revolutionized many domains, e.g., image recognition, speech recognition, speech synthesis, and knowledge discovery (Krizhevsky et al, 2012; LeCun et al, 2012; Schmidhuber, 2015; LeCun et al, 2015; Van Den Oord et al, 2016)
  • Due to their ability to naturally learn from structured data and exhibit superior performance, they are increasingly used in practical applications and critical decision processes, such as novel knowledge discovery techniques, autonomous driving or medical image analysis.
  • In order to evaluate these methods, the authors present iNNvestigate which provides a common interface to a variety of analysis methods
Highlights
  • In recent years deep neural networks have revolutionized many domains, e.g., image recognition, speech recognition, speech synthesis, and knowledge discovery (Krizhevsky et al, 2012; LeCun et al, 2012; Schmidhuber, 2015; LeCun et al, 2015; Van Den Oord et al, 2016). Due to their ability to naturally learn from structured data and exhibit superior performance, they are increasingly used in practical applications and critical decision processes, such as novel knowledge discovery techniques, autonomous driving or medical image analysis
  • In order to evaluate these methods, we present iNNvestigate which provides a common interface to a variety of analysis methods
  • We have presented iNNvestigate, a library that makes it easier to analyze neural networks’ predictions and to compare different analysis methods
  • This is done by providing a common interface and implementations for many analysis methods as well as making tools for training and comparing methods available. In particular it contains reference implementations for many methods (PatternNet, PatternAttribution, LRP) and example application for a large number of state-of-the-art applications. We expect that this library will support the field of analyzing machine learning and facilitate research using neural networks in domains such as drug design or medical image analysis
Methods
  • At publication time the following algorithms are supported: Gradient Saliency Map, SmoothGrad, IntegratedGradients, Deconvnet, GuidedBackprop, PatternNet and PatternAttribution, DeepTaylor, and LRP including LRP-Z, -Epsilon, -AlphaBeta.
  • Current related work is limited to gradient- and perturbation-based methods (Kotikalapudi and contributors, 2017; Ancona et al, 2018) or focuses on a single algorithm (E.g., Lundberg and Lee, 2017; Ribeiro et al, 2016).
  • The authors intend to extend this selection and invite the community to contribute implementations as new methods emerge.
Conclusion
  • The authors have presented iNNvestigate, a library that makes it easier to analyze neural networks’ predictions and to compare different analysis methods.
  • In particular it contains reference implementations for many methods (PatternNet, PatternAttribution, LRP) and example application for a large number of state-of-the-art applications
  • The authors expect that this library will support the field of analyzing machine learning and facilitate research using neural networks in domains such as drug design or medical image analysis
Summary
  • Introduction:

    In recent years deep neural networks have revolutionized many domains, e.g., image recognition, speech recognition, speech synthesis, and knowledge discovery (Krizhevsky et al, 2012; LeCun et al, 2012; Schmidhuber, 2015; LeCun et al, 2015; Van Den Oord et al, 2016)
  • Due to their ability to naturally learn from structured data and exhibit superior performance, they are increasingly used in practical applications and critical decision processes, such as novel knowledge discovery techniques, autonomous driving or medical image analysis.
  • In order to evaluate these methods, the authors present iNNvestigate which provides a common interface to a variety of analysis methods
  • Methods:

    At publication time the following algorithms are supported: Gradient Saliency Map, SmoothGrad, IntegratedGradients, Deconvnet, GuidedBackprop, PatternNet and PatternAttribution, DeepTaylor, and LRP including LRP-Z, -Epsilon, -AlphaBeta.
  • Current related work is limited to gradient- and perturbation-based methods (Kotikalapudi and contributors, 2017; Ancona et al, 2018) or focuses on a single algorithm (E.g., Lundberg and Lee, 2017; Ribeiro et al, 2016).
  • The authors intend to extend this selection and invite the community to contribute implementations as new methods emerge.
  • Conclusion:

    The authors have presented iNNvestigate, a library that makes it easier to analyze neural networks’ predictions and to compare different analysis methods.
  • In particular it contains reference implementations for many methods (PatternNet, PatternAttribution, LRP) and example application for a large number of state-of-the-art applications
  • The authors expect that this library will support the field of analyzing machine learning and facilitate research using neural networks in domains such as drug design or medical image analysis
Funding
  • This work was supported by the Federal Ministry of Education and Research (BMBF) for the Berlin Big Data Center BBDC (01IS14013A)
  • Additional support was provided by the BK21 program funded by Korean National Research Foundation grant (No 2012-005741) and the Institute for Information & Communications Technology Promotion (IITP) grant funded by the Korea government (no. 2017-0-00451, No 2017-0-01779)
Reference
  • Martın Abadi, Paul Barham, Jianmin Chen, Zhifeng Chen, Andy Davis, Jeffrey Dean, Matthieu Devin, Sanjay Ghemawat, Geoffrey Irving, Michael Isard, et al. Tensorflow: a system for large-scale machine learning. Proceedings of the 11th USENIX Symposium on Operating Systems Design and Implementation, 16:265–283, 2016.
    Google ScholarLocate open access versionFindings
  • Maximilian Alber. Software and application patterns for explanation methods. In Interpretable AI: Interpreting, Explaining and Visualizing Deep Learning. Springer, 2019.
    Google ScholarFindings
  • Marco Ancona, Enea Ceolini, Cengiz Oztireli, and Markus Gross. Towards better understanding of gradient-based attribution methods for deep neural networks. In International Conference on Learning Representations, 2018.
    Google ScholarLocate open access versionFindings
  • Sebastian Bach, Alexander Binder, Gregoire Montavon, Frederick Klauschen, Klaus-Robert Muller, and Wojciech Samek. On pixel-wise explanations for non-linear classifier decisions by layer-wise relevance propagation. PLOS ONE, 10(7):1–46, 2015.
    Google ScholarLocate open access versionFindings
  • David Baehrens, Timon Schroeter, Stefan Harmeling, Motoaki Kawanabe, Katja Hansen, and Klaus-Robert Muller. How to explain individual classification decisions. Journal of Machine Learning Research, 11:1803–1831, 2010.
    Google ScholarLocate open access versionFindings
  • Francois Chollet et al. Keras. https://github.com/fchollet/keras, 2015.
    Findings
  • Jia Deng, Wei Dong, Richard Socher, Li-Jia Li, Kai Li, and Li Fei-Fei. Imagenet: A largescale hierarchical image database. In 2009 IEEE Conference on Computer Vision and Pattern Recognition, pages 248–255, 2009.
    Google ScholarLocate open access versionFindings
  • Kaiming He, Xiangyu Zhang, Shaoqing Ren, and Jian Sun. Deep residual learning for image recognition. In 2016 IEEE Conference on Computer Vision and Pattern Recognition, pages 770–778, 2016.
    Google ScholarLocate open access versionFindings
  • Gao Huang, Zhuang Liu, Laurens van der Maaten, and Kilian Q. Weinberger. Densely connected convolutional networks. In 2017 IEEE Conference on Computer Vision and Pattern Recognition, pages 2261–2269, 2017.
    Google ScholarLocate open access versionFindings
  • Pieter-Jan Kindermans, Sara Hooker, Julius Adebayo, Maximilian Alber, Kristof T Schutt, Sven Dahne, Dumitru Erhan, and Been Kim. The (un)reliability of saliency methods. Neural Information Processing Systems 2017 - Interpreting, Explaining and Visualizing Deep Learning - Now what? workshop, 2017.
    Google ScholarLocate open access versionFindings
  • Pieter-Jan Kindermans, Kristof T Schutt, Maximilian Alber, Klaus-Robert Muller, Dumitru Erhan, Been Kim, and Sven Dahne. Learning how to explain neural networks: PatternNet and PatternAttribution. In International Conference on Learning Representations, 2018.
    Google ScholarLocate open access versionFindings
  • Raghavendra Kotikalapudi and contributors. keras-vis. https://github.com/raghakot/keras-vis, 2017.
    Findings
  • Alex Krizhevsky, Ilya Sutskever, and Geoffrey E Hinton. Imagenet classification with deep convolutional neural networks. In Advances in Neural Information Processing Systems 25, pages 1097–1105, 2012.
    Google ScholarLocate open access versionFindings
  • Sebastian Lapuschkin, Alexander Binder, Gregoire Montavon, Klaus-Robert Muller, and Wojciech Samek. Analyzing classifiers: Fisher vectors and deep neural networks. In 2016 IEEE Conference on Computer Vision and Pattern Recognition, pages 2912–2920, 2016a.
    Google ScholarLocate open access versionFindings
  • Sebastian Lapuschkin, Alexander Binder, Gregoire Montavon, Klaus-Robert Muller, and Wojciech Samek. The layer-wise relevance propagation toolbox for artificial neural networks. Journal of Machine Learning Research, 17:3938–3942, 2016b.
    Google ScholarLocate open access versionFindings
  • Sebastian Lapuschkin, Stephan Waldchen, Alexander Binder, Gregoire Montavon, Wojciech Samek, and Klaus-Robert Muller. Unmasking clever hans predictors and assessing what machines really learn. Nature communcations, 10:ID: 1096, 2019.
    Google ScholarLocate open access versionFindings
  • Yann A LeCun, Leon Bottou, Genevieve B Orr, and Klaus-Robert Muller. Efficient backprop. In Neural networks: Tricks of the trade, pages 9–48. 2012.
    Google ScholarLocate open access versionFindings
  • Yann A LeCun, Yoshua Bengio, and Geoffrey E Hinton. Deep learning. Nature, 521(7553): 436–444, 2015.
    Google ScholarLocate open access versionFindings
  • Scott M Lundberg and Su-In Lee. A unified approach to interpreting model predictions. In Advances in Neural Information Processing Systems 30, pages 4765–4774, 2017.
    Google ScholarLocate open access versionFindings
  • Gregoire Montavon, Matthias Rupp, Vivekanand Gobre, Alvaro Vazquez-Mayagoitia, Katja Hansen, Alexandre Tkatchenko, Klaus-Robert Muller, and Anatole Von Lilienfeld. Machine learning of molecular electronic properties in chemical compound space. New Journal of Physics, 15(9):ID: 095003, 2013.
    Google ScholarLocate open access versionFindings
  • Gregoire Montavon, Sebastian Bach, Alexander Binder, Wojciech Samek, and Klaus-Robert Muller. Explaining nonlinear classification decisions with deep taylor decomposition. Pattern Recognition, 65:211–222, 2017.
    Google ScholarLocate open access versionFindings
  • Gregoire Montavon, Wojciech Samek, and Klaus-Robert Muller. Methods for interpreting and understanding deep neural networks. Digital Signal Processing, 73:1–15, 2018.
    Google ScholarLocate open access versionFindings
  • Marco Tulio Ribeiro, Sameer Singh, and Carlos Guestrin. ”why should i trust you?”: Explaining the predictions of any classifier. In Proceedings of the 22th ACM SIGKDD International Conference on Knowledge Discovery and Data Mining, pages 1135–1144, 2016.
    Google ScholarLocate open access versionFindings
  • Wojciech Samek, Alexander Binder, Gregoire Montavon, Sebastian Lapuschkin, and KlausRobert Muller. Evaluating the visualization of what a deep neural network has learned. IEEE Transactions on Neural Networks and Learning Systems, 28(11):2660–2673, 2017.
    Google ScholarLocate open access versionFindings
  • Jurgen Schmidhuber. Deep learning in neural networks: An overview. Neural networks, 61: 85–117, 2015.
    Google ScholarLocate open access versionFindings
  • Kristof T Schutt, Farhad Arbabzadah, Stefan Chmiela, Klaus-Robert Muller, and Alexandre Tkatchenko. Quantum-chemical insights from deep tensor neural networks. Nature Communications, 8:ID: 13890, 2017.
    Google ScholarLocate open access versionFindings
  • Karen Simonyan and Andrew Zisserman. Very deep convolutional networks for large-scale image recognition. arXiv preprint arXiv:1409.1556, 2014.
    Findings
  • Karen Simonyan, Andrea Vedaldi, and Andrew Zisserman. Deep inside convolutional networks: Visualising image classification models and saliency maps. arXiv preprint arXiv:1312.6034, 2013.
    Findings
  • Daniel Smilkov, Nikhil Thorat, Been Kim, Fernanda Viegas, and Martin Wattenberg. Smoothgrad: Removing noise by adding noise. International Conference on Machine Learning 2017 - Workshop on Visualization for Deep Learning, 2017.
    Google ScholarLocate open access versionFindings
  • Jost T Springenberg, Alexey Dosovitskiy, Thomas Brox, and Martin Riedmiller. Striving for simplicity: The all convolutional net. In International Conference on Learning Representations - Workshop track, 2015.
    Google ScholarLocate open access versionFindings
  • Mukund Sundararajan, Ankur Taly, and Qiqi Yan. Axiomatic attribution for deep networks. In Proceedings of the 34th International Conference on Machine Learning, pages 3319– 3328, 2017.
    Google ScholarLocate open access versionFindings
  • Christian Szegedy, Vincent Vanhoucke, Sergey Ioffe, Jon Shlens, and Zbigniew Wojna. Rethinking the inception architecture for computer vision. In 2016 IEEE Conference on Computer Vision and Pattern Recognition, pages 2818–2826, 2016.
    Google ScholarLocate open access versionFindings
  • Christian Szegedy, Sergey Ioffe, Vincent Vanhoucke, and Alexander A Alemi. Inception-v4, inception-resnet and the impact of residual connections on learning. In Proceedings of the 31st AAAI Conference on Artificial Intelligence (AAAI), pages 4278–4284, 2017.
    Google ScholarLocate open access versionFindings
  • Aaron Van Den Oord, Sander Dieleman, Heiga Zen, Karen Simonyan, Oriol Vinyals, Alex Graves, Nal Kalchbrenner, Andrew Senior, and Koray Kavukcuoglu. Wavenet: A generative model for raw audio. arXiv preprint arXiv:1609.03499, 2016.
    Findings
  • Matthew D Zeiler and Rob Fergus. Visualizing and understanding convolutional networks. In Proceedings of the 2014 European Conference on Computer Vision, pages 818–833, 2014.
    Google ScholarLocate open access versionFindings
  • Barret Zoph, Vijay Vasudevan, Jonathon Shlens, and Quoc V Le. Learning transferable architectures for scalable image recognition. 2018 IEEE Conference on Computer Vision and Pattern Recognition, pages 8697–8710, 2018.
    Google ScholarLocate open access versionFindings
Full Text
Your rating :
0

 

Tags
Comments