Assessing Capsule Networks With Biased Data

Machine learning based methods achieves impressive results in object classification and detection. Utilizing representative data of the visual world during the training phase is crucial to achieve good performance with such data driven approaches. However, it not always possible to access bias-free datasets thus, robustness to biased data is a desirable property for a learning system. Capsule Networks have been introduced recently and their tolerance to biased data has received little attention. This paper aims to fill this gap and proposes two experimental scenarios to assess the tolerance to imbalanced training data and to determine the generalization performance of a model with unfamiliar affine transformations of the images. This paper assesses dynamic routing and EM routing based Capsule Networks and proposes a comparison with Convolutional Neural Networks in the two tested scenarios. The presented results provide new insights into the behaviour of capsule networks.

[1]  David Masko,et al.  The Impact of Imbalanced Training Data for Convolutional Neural Networks , 2015 .

[2]  Ales Leonardis,et al.  Evaluating Deep Convolutional Neural Networks for Material Classification , 2017, VISIGRAPP.

[3]  Gregory Cohen,et al.  EMNIST: an extension of MNIST to handwritten letters , 2017, CVPR 2017.

[4]  Martín Abadi,et al.  TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems , 2016, ArXiv.

[5]  Radu State,et al.  Impact of Biases in Big Data , 2018, ESANN.

[6]  Barbara Caputo,et al.  A Deeper Look at Dataset Bias , 2015, Domain Adaptation in Computer Vision Applications.

[7]  Geoffrey E. Hinton,et al.  Matrix capsules with EM routing , 2018, ICLR.

[8]  Torsten Sattler,et al.  Quad-Networks: Unsupervised Learning to Rank for Interest Point Detection , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[9]  Alexei A. Efros,et al.  Unbiased look at dataset bias , 2011, CVPR 2011.

[10]  Bernhard Egger,et al.  Empirically Analyzing the Effect of Dataset Biases on Deep Face Recognition Systems , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition Workshops (CVPRW).

[11]  Andrew Gritsevskiy,et al.  Capsule networks for low-data transfer learning , 2018, ArXiv.

[12]  Rohan Doshi,et al.  Pushing the Limits of Capsule Networks , 2018 .

[13]  Josef Sivic,et al.  NetVLAD: CNN Architecture for Weakly Supervised Place Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[14]  Geoffrey E. Hinton,et al.  Dynamic Routing Between Capsules , 2017, NIPS.

[15]  Guigang Zhang,et al.  Deep Learning , 2016, Int. J. Semantic Comput..

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

[17]  Alex Krizhevsky,et al.  Learning Multiple Layers of Features from Tiny Images , 2009 .

[18]  Alexei A. Efros,et al.  Undoing the Damage of Dataset Bias , 2012, ECCV.

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

[20]  Yoshua Bengio,et al.  Gradient-based learning applied to document recognition , 1998, Proc. IEEE.

[21]  Yang Jin,et al.  Capsule Network Performance on Complex Data , 2017, ArXiv.

[22]  Cesare Furlanello,et al.  A Comparison of MCC and CEN Error Measures in Multi-Class Prediction , 2010, PloS one.

[23]  Josephine Sarpong Akosa,et al.  Predictive Accuracy : A Misleading Performance Measure for Highly Imbalanced Data , 2017 .

[24]  Gregory Cohen,et al.  EMNIST: Extending MNIST to handwritten letters , 2017, 2017 International Joint Conference on Neural Networks (IJCNN).

[25]  Geoffrey E. Hinton,et al.  Transforming Auto-Encoders , 2011, ICANN.

[26]  Stefano Soatto,et al.  An Empirical Evaluation of Current Convolutional Architectures’ Ability to Manage Nuisance Location and Scale Variability , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).