Deep Self-Taught Learning for Handwritten Character Recognition

by   Frédéric Bastien, et al.

Recent theoretical and empirical work in statistical machine learning has demonstrated the importance of learning algorithms for deep architectures, i.e., function classes obtained by composing multiple non-linear transformations. Self-taught learning (exploiting unlabeled examples or examples from other distributions) has already been applied to deep learners, but mostly to show the advantage of unlabeled examples. Here we explore the advantage brought by out-of-distribution examples. For this purpose we developed a powerful generator of stochastic variations and noise processes for character images, including not only affine transformations but also slant, local elastic deformations, changes in thickness, background images, grey level changes, contrast, occlusion, and various types of noise. The out-of-distribution examples are obtained from these highly distorted images or by including examples of object classes different from those in the target test set. We show that deep learners benefit more from out-of-distribution examples than a corresponding shallow learner, at least in the area of handwritten character recognition. In fact, we show that they beat previously published results and reach human-level performance on both handwritten digit classification and 62-class handwritten character recognition.


page 4

page 5

page 6


Classification Of Gradient Change Features Using MLP For Handwritten Character Recognition

A novel, generic scheme for off-line handwritten English alphabets chara...

Performance Comparison of SVM and ANN for Handwritten Devnagari Character Recognition

Classification methods based on learning from examples have been widely ...

Pixel-level Reconstruction and Classification for Noisy Handwritten Bangla Characters

Classification techniques for images of handwritten characters are susce...

Development of Comprehensive Devnagari Numeral and Character Database for Offline Handwritten Character Recognition

In handwritten character recognition, benchmark database plays an import...

A Topological Code for Plane Images

It is proposed a new code for contours of plane images. This code was ap...

A GA Based approach for selection of local features for recognition of handwritten Bangla numerals

Soft computing approaches are mainly designed to address the real world ...

Please sign up or login with your details

Forgot password? Click here to reset