Classification involves obtaining a sample and determining to which class the sample belongs (from a finite set of classes). The simplest form of classification is binary classification, where you must determine whether it is of class A or class B. The task ultimately boils down to determining identifying features of each class that are distinct from each other, then using those features to define a selection or classification policy. Here, we will explore different ways to generate or discover the identifying features, also known as the feature descriptor or feature vector of the sample.

This set of learning modules will go a little backwards relative to how our understanding has evolved regarding classification. It starts with an unsupervised method for performing feature descriptor estimation, where machine learning tools are used to come up with the identifying features. Later methods to be explored are slightly more engineered solutions that have been shown to work well for some problems.

/* (2) bag-of-words classifier: http://people.csail.mit.edu/fergus/iccv2005/bagwords.html They were short courses on ICCV 2005. */

/* Andrew Ng */

The classical artificial intelligence/machine learning example of classification is digits classification. Can a computer learn to recognize the 10 digits 0 through 9? Clearly this question was of great utility to the United State Post Office as well as to many banks (for automatically processing cashed checks). Most likely the first person to really solve this well is now fairly wealthy (can you figure out who this person might be?).

This set of activities comes courtesy of Prof. Andrew Ng at Stanford university. We will be using his tutorials , which you will read as you go through this learning module.

Weekly activities (one bullet per week)

1. Implement the 'sparse autoencoder' section.
2. Get 'Vectorized implementation', 'Preprocessing: PCA and Whitening' and 'Softmax Regression' sections done.
3. Implement 'Self-Taught Learning and Unsupervised Feature Learning' section for digit classification.

Classification on digits using DEEP neural networks

1. Once you finish module 2, you should have a nice classification system on digit with 98% accuracy.
2. Now, the advanced task would require you to extend it to be deep neural networks
3. In this module, based on your module 1, please continue to work on 'Building Deep Networks for Classification'

Classification on characters using DEEP neural networks

1. Now you have a very nice classification system on digit. Can we classify on other similar data?
2. In module 3, take your neural networks and train it with characters instead of digits. For characters, we would like to use Omniglot dataset.
3. Report your results and comparison. Also please test if the sparse code learned from digits can be used to test on characters?

/* Other related material, maybe more classic. */

ECE4580 Learning Modules