How to Pick the Best Source Data? Measuring Transferability for Heterogeneous Domains

Given a set of source data with pre-trained classification models, how can we fast and accurately select the most useful source data to improve the performance of a target task? We address the problem of measuring transferability for heterogeneous domains, where the source and the target data have different feature spaces and distributions. We propose Transmeter, a novel method to efficiently and accurately measure transferability of two datasets. Transmeter utilizes a pre-trained source classifier and a reconstruction loss to increase its efficiency and performance. Furthermore, Transmeter uses feature transformation layers, label-wise discriminators, and a mean distance loss to learn common representations for source and target domains. As a result, Transmeter and its variant give the most accurate performance in measuring transferability, while giving comparable running times compared to those of competitors.

[1]  Jaime G. Carbonell,et al.  Characterizing and Avoiding Negative Transfer , 2018, 2019 IEEE/CVF Conference on Computer Vision and Pattern Recognition (CVPR).

[2]  Ivor W. Tsang,et al.  Learning with Augmented Features for Heterogeneous Domain Adaptation , 2012, ICML.

[3]  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.

[4]  M. Mead,et al.  Cybernetics , 1953, The Yale Journal of Biology and Medicine.

[5]  Philip S. Yu,et al.  2014 IEEE International Conference on Data Mining , 2014 .

[6]  Jing Gao,et al.  On handling negative transfer and imbalanced distributions in multiple source transfer learning , 2014, SDM.

[7]  Athanasios V. Vasilakos,et al.  Big data: From beginning to future , 2016, Int. J. Inf. Manag..

[8]  François Laviolette,et al.  Domain-Adversarial Training of Neural Networks , 2015, J. Mach. Learn. Res..

[9]  Yi Yao,et al.  Boosting for transfer learning with multiple sources , 2010, 2010 IEEE Computer Society Conference on Computer Vision and Pattern Recognition.

[10]  Ivor W. Tsang,et al.  Heterogeneous Domain Adaptation for Multiple Classes , 2014, AISTATS.

[11]  Dong Xu,et al.  Exploiting web images for event recognition in consumer videos: A multiple source domain adaptation approach , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[12]  Qiang Yang,et al.  A Survey on Transfer Learning , 2010, IEEE Transactions on Knowledge and Data Engineering.

[13]  Michael K. Ng,et al.  Learning Discriminative Correlation Subspace for Heterogeneous Domain Adaptation , 2017, IJCAI.

[14]  Helmut Krcmar,et al.  Big Data , 2014, Wirtschaftsinf..

[15]  Hal Daumé,et al.  Frustratingly Easy Domain Adaptation , 2007, ACL.

[16]  Trevor Darrell,et al.  What you saw is not what you get: Domain adaptation using asymmetric kernel transforms , 2011, CVPR 2011.

[17]  Ivor W. Tsang,et al.  Learning With Augmented Features for Supervised and Semi-Supervised Heterogeneous Domain Adaptation , 2014, IEEE Transactions on Pattern Analysis and Machine Intelligence.

[18]  Ivor W. Tsang,et al.  Combating Negative Transfer From Predictive Distribution Differences , 2013, IEEE Transactions on Cybernetics.

[19]  Qiang Yang,et al.  Cross-domain sentiment classification via spectral feature alignment , 2010, WWW '10.

[20]  Taghi M. Khoshgoftaar,et al.  A survey of transfer learning , 2016, Journal of Big Data.

[21]  Philip S. Yu,et al.  Transfer across Completely Different Feature Spaces via Spectral Embedding , 2013, IEEE Transactions on Knowledge and Data Engineering.

[22]  De-Chuan Zhan,et al.  Distance Metric Facilitated Transportation between Heterogeneous Domains , 2018, IJCAI.

[23]  Philip S. Yu,et al.  Transfer Learning on Heterogenous Feature Spaces via Spectral Transformation , 2010, 2010 IEEE International Conference on Data Mining.

[24]  Chang Wang,et al.  Heterogeneous Domain Adaptation Using Manifold Alignment , 2011, IJCAI.

[25]  Yuan Shi,et al.  Geodesic flow kernel for unsupervised domain adaptation , 2012, 2012 IEEE Conference on Computer Vision and Pattern Recognition.

[26]  Peter A. Flach,et al.  Proceedings of the 28th International Conference on Machine Learning , 2011 .

[27]  Jianmin Wang,et al.  Partial Transfer Learning with Selective Adversarial Networks , 2017, 2018 IEEE/CVF Conference on Computer Vision and Pattern Recognition.