000218496 001__ 218496
000218496 005__ 20190812205914.0
000218496 020__ $$a978-1-4673-9961-6
000218496 022__ $$a1522-4880
000218496 02470 $$2ISI$$a000390782003139
000218496 037__ $$aCONF
000218496 245__ $$aAdaptive data augmentation for image classification
000218496 269__ $$a2016
000218496 260__ $$bIEEE$$c2016$$aNew York
000218496 300__ $$a5
000218496 336__ $$aConference Papers
000218496 490__ $$aIEEE International Conference on Image Processing ICIP
000218496 520__ $$aData augmentation is the process of generating samples by transforming training data, with the target of improving the accuracy and robustness of classifiers. In this paper, we propose a new automatic and adaptive algorithm for choosing the transformations of the samples used in data augmentation. Specifically, for each sample, our main idea is to seek a small transformation that yields maximal classification loss on the transformed sample. We employ a trust-region optimization strategy, which consists of solving a sequence of linear programs. Our data augmentation scheme is then integrated into a Stochastic Gradient Descent algorithm for training deep neural networks. We perform experiments on two datasets, and show that that the proposed scheme outperforms random data augmentation algorithms in terms of accuracy and robustness, while yielding comparable or superior results with respect to existing selective sampling approaches.
000218496 6531_ $$aData augmentation
000218496 6531_ $$aTransformation invariance
000218496 6531_ $$aRobustness
000218496 700__ $$0246320$$g203034$$aFawzi, Alhussein
000218496 700__ $$aSamulowitz, Horst
000218496 700__ $$aTuraga, Deepak
000218496 700__ $$0241061$$g101475$$aFrossard, Pascal
000218496 7112_ $$a23rd IEEE International Conference on Image Processing (ICIP)
000218496 773__ $$t2016 Ieee International Conference On Image Processing (Icip)$$q3688-3692
000218496 8564_ $$zn/a$$yn/a$$uhttps://infoscience.epfl.ch/record/218496/files/ICIP_CAMERAREADY_2715.pdf$$s229901
000218496 909C0 $$xU10851$$pLTS4$$0252393
000218496 909CO $$ooai:infoscience.tind.io:218496$$qGLOBAL_SET$$pconf$$pSTI
000218496 917Z8 $$x203034
000218496 917Z8 $$x203034
000218496 917Z8 $$x203034
000218496 917Z8 $$x101475
000218496 917Z8 $$x144315
000218496 937__ $$aEPFL-CONF-218496
000218496 973__ $$rREVIEWED$$sPUBLISHED$$aEPFL
000218496 980__ $$aCONF