scholarly journals Deep, Big, Simple Neural Nets for Handwritten Digit Recognition

2010 ◽  
Vol 22 (12) ◽  
pp. 3207-3220 ◽  
Author(s):  
Dan Claudiu Cireşan ◽  
Ueli Meier ◽  
Luca Maria Gambardella ◽  
Jürgen Schmidhuber
Keyword(s):  
Error Rate ◽  
Neural Nets ◽  
Multilayer Perceptrons ◽  
Handwritten Digit Recognition ◽  
Digit Recognition ◽  
Training Images ◽  
Speed Up ◽  
Handwritten Digit

Good old online backpropagation for plain multilayer perceptrons yields a very low 0.35% error rate on the MNIST handwritten digits benchmark. All we need to achieve this best result so far are many hidden layers, many neurons per layer, numerous deformed training images to avoid overfitting, and graphics cards to greatly speed up learning.

scholarly journals On the Brittleness of Handwritten Digit Recognition Models

2012 ◽  
Vol 2012 ◽  
pp. 1-10 ◽  
Author(s):  
Alexander K. Seewald
Keyword(s):  
Error Rate ◽  
Maximum Error ◽  
Error Rates ◽  
Classification Systems ◽  
Feature Representation ◽  
Training Data ◽  
Handwritten Digit Recognition ◽  
Digit Recognition ◽  
Handwritten Digit ◽  
The Relationship

Handwritten digit recognition is an important benchmark task in computer vision. Learning algorithms and feature representations which offer excellent performance for this task have been known for some time. Here, we focus on two major practical considerations: the relationship between the the amount of training data and error rate (corresponding to the effort to collect training data to build a model with a given maximum error rate) and the transferability of models' expertise between different datasets (corresponding to the usefulness for general handwritten digit recognition). While the relationship between amount of training data and error rate is very stable and to some extent independent of the specific dataset used—only the classifier and feature representation have significant effect—it has proven to be impossible to transfer low error rates on one or two pooled datasets to similarly low error rates on another dataset. We have called this weakness brittleness, inspired by an old Artificial Intelligence term that means the same thing. This weakness may be a general weakness of trained image classification systems.

Sign in / Sign up

Export Citation Format

Share Document