Table of results for MNIST

This is a table documenting some of the best results some paper obtained in MNIST dataset.

Results shown indicates the error obtained by training on all 60,000 samples and testing on 10,000 samples.

1. What is the Best Multi-Stage Architecture for Object Recognition? (2009)
   0.53%
2. Simple Methods for High-Performance Digit Recognition Based on Sparse Coding (2008)
   0.59%
   
3. Efficient Learning of Sparse Representations with an Energy-Based Model (2006)
   0.6%
   Additional Info: If train with 60k + distortions, the error is 0.39%
   
4. Unsupervised learning of invariant feature hierarchies with applications to object recognition (2007)
   0.62%
   Additional Info: Supervised training from random initial conditions
   
5. Convolutional Deep Belief Networks for Scalable Unsupervised Learning of Hierarchical Representations (2009)
   0.82%
6. Deep Boltzmann Machines (2009)
   0.95%
7. CS81: Learning words with Deep Belief Networks (2008)
   1.12%
   
8. Reducing the dimensionality of data with neural networks (2006)
   1.2%
9. Deep learning via semi-supervised embedding (2008)
   1.5%
   

Table of results for Caltech 256

This is a table documenting some of the best results some paper obtained in Caltech-256 dataset.

1. Image Classification using Random Forests and Ferns (2007)
   Cited 51 times. 45.3%