Adapting Convolutional Neural Networks for Geographical Domain Shift

We present the winning solution for the Inclusive Images Competition organized as part of the Conference on Neural Information Processing Systems (NeurIPS 2018) Competition Track. The competition was organized to study ways to cope with domain shift in image processing, specifically geographical shift: the training and two test sets in the competition had different geographical distributions. Our solution has proven to be relatively straightforward and simple: it is an ensemble of several CNNs where only the last layer is fine-tuned with the help of a small labeled set of tuning labels made available by the organizers. We believe that while domain shift remains a formidable problem, our approach opens up new possibilities for alleviating this problem in practice, where small labeled datasets from the target domain are usually either available or can be obtained and labeled cheaply.

[1]  Trevor Darrell,et al.  DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition , 2013, ICML.

[2]  Vijay Vasudevan,et al.  Learning Transferable Architectures for Scalable Image Recognition , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[3]  Kilian Q. Weinberger,et al.  Densely Connected Convolutional Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[4]  Fei-Fei Li,et al.  ImageNet: A large-scale hierarchical image database , 2009, 2009 IEEE Conference on Computer Vision and Pattern Recognition.

[5]  Taesung Park,et al.  CyCADA: Cycle-Consistent Adversarial Domain Adaptation , 2017, ICML.

[6]  Ivan Laptev,et al.  Learning and Transferring Mid-level Image Representations Using Convolutional Neural Networks , 2014, 2014 IEEE Conference on Computer Vision and Pattern Recognition.

[7]  Alexei A. Efros,et al.  Image-to-Image Translation with Conditional Adversarial Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[8]  Koby Crammer,et al.  Learning from Multiple Sources , 2006, NIPS.

[9]  Ming-Yu Liu,et al.  Coupled Generative Adversarial Networks , 2016, NIPS.

[10]  Stephen Cox,et al.  RecNorm: Simultaneous Normalisation and Classification Applied to Speech Recognition , 1990, NIPS.

[11]  Jitendra Malik,et al.  Are All Training Examples Created Equal? An Empirical Study , 2018, ArXiv.

[12]  Trevor Darrell,et al.  Adapting Visual Category Models to New Domains , 2010, ECCV.

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

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

[15]  Mengjie Zhang,et al.  Deep Reconstruction-Classification Networks for Unsupervised Domain Adaptation , 2016, ECCV.

[16]  Gang Sun,et al.  Squeeze-and-Excitation Networks , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.

[17]  Koby Crammer,et al.  Learning Bounds for Domain Adaptation , 2007, NIPS.

[18]  Alexei A. Efros,et al.  Unpaired Image-to-Image Translation Using Cycle-Consistent Adversarial Networks , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[19]  James Zou,et al.  AI can be sexist and racist — it’s time to make it fair , 2018, Nature.

[20]  Koby Crammer,et al.  A theory of learning from different domains , 2010, Machine Learning.

[21]  Zhiguo Cao,et al.  When Unsupervised Domain Adaptation Meets Tensor Representations , 2017, 2017 IEEE International Conference on Computer Vision (ICCV).

[22]  Jordi Pont-Tuset,et al.  The Open Images Dataset V4 , 2018, International Journal of Computer Vision.

[23]  Dumitru Erhan,et al.  Unsupervised Pixel-Level Domain Adaptation with Generative Adversarial Networks , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[24]  Tomas Pfister,et al.  Learning from Simulated and Unsupervised Images through Adversarial Training , 2016, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[25]  Victor S. Lempitsky,et al.  Unsupervised Domain Adaptation by Backpropagation , 2014, ICML.

[26]  Li Fei-Fei,et al.  Progressive Neural Architecture Search , 2017, ECCV.

[27]  Alexandr A. Kalinin,et al.  Albumentations: fast and flexible image augmentations , 2018, Inf..

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

[29]  D. Sculley,et al.  No Classification without Representation: Assessing Geodiversity Issues in Open Data Sets for the Developing World , 2017, 1711.08536.

[30]  Trevor Darrell,et al.  Adversarial Discriminative Domain Adaptation , 2017, 2017 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[31]  Mei Wang,et al.  Deep Visual Domain Adaptation: A Survey , 2018, Neurocomputing.

[32]  Rama Chellappa,et al.  Visual Domain Adaptation: A survey of recent advances , 2015, IEEE Signal Processing Magazine.

[33]  Philip Ogunbona,et al.  Cross-Dataset Recognition: A Survey , 2017, ArXiv.

[34]  Jian Sun,et al.  Deep Residual Learning for Image Recognition , 2015, 2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR).

[35]  Stefano Ermon,et al.  A DIRT-T Approach to Unsupervised Domain Adaptation , 2018, ICLR.

[36]  Trevor Darrell,et al.  Simultaneous Deep Transfer Across Domains and Tasks , 2015, ICCV.

[37]  Namil Kim,et al.  Pixel-Level Domain Transfer , 2016, ECCV.

[38]  Quoc V. Le,et al.  Don't Decay the Learning Rate, Increase the Batch Size , 2017, ICLR.