Service interruption on Monday 11 July from 12:30 to 13:00: all the sites of the CCSD (HAL, EpiSciences, SciencesConf, AureHAL) will be inaccessible (network hardware connection).
Abstract : Deep learning is promising approach to extract useful nonlinear representations of data. However, it is usually applied with large training sets, which are not always available in practical tasks. In this paper, we consider stacked autoencoders with logistic regression as the classification layer and study their usefulness for the task of image categorization depending on the size of training sets. Hand-crafted image descriptors are proposed and used for training autoencoders in addition to pixel-level features. New multi-column architecture for autoencoders is also proposed. Conducted experiments showed that useful nonlinear features can be learnt by (stacked) autoencoders only using large training sets, but they can yield positive results due to redundancy reduction also on small training sets. Practically useful results (9.1% error rate for 6 classes) were achieved only using hand-crafted features on the training set containing 4800 images.
https://hal.inria.fr/hal-01391322 Contributor : Hal IfipConnect in order to contact the contributor Submitted on : Thursday, November 3, 2016 - 10:58:38 AM Last modification on : Thursday, March 5, 2020 - 5:41:14 PM Long-term archiving on: : Saturday, February 4, 2017 - 1:16:18 PM
Alexey Potapov, Vita Batishcheva, Maxim Peterson. Limited Generalization Capabilities of Autoencoders with Logistic Regression on Training Sets of Small Sizes. 10th IFIP International Conference on Artificial Intelligence Applications and Innovations (AIAI), Sep 2014, Rhodes, Greece. pp.256-264, ⟨10.1007/978-3-662-44654-6_25⟩. ⟨hal-01391322⟩