The competitive MNIST handwritten digit recognition benchmark has a long history of broken records since 1998. The most recent advancement by others dates back 8 years (error rate 0.4 old on-line back-propagation for plain multi-layer perceptrons yields a very low 0.35% error rate on the MNIST handwritten digits benchmark with a single MLP and 0.31% with a committee of seven MLP. All we need to achieve this until 2011 best result are many hidden layers, many neurons per layer, numerous deformed training images to avoid overfitting, and graphics cards to greatly speed up learning. © Springer-Verlag Berlin Heidelberg 2012.
|Original language||English (US)|
|Number of pages||18|
|Journal||Lecture Notes in Computer Science (including subseries Lecture Notes in Artificial Intelligence and Lecture Notes in Bioinformatics)|
|Volume||7700 LECTURE NO|
|State||Published - Jan 1 2012|