Restricted Boltzmann Machines

March 28, 2011 at 11:31 am (Computer Science) (, , , , , , , , , , , , )

So for a computational neuroscience project, I’ve been working at implementing a type of neural network called a restricted Boltzmann machine (RBM). It’s a type of unsupervised learning consisting of a visible and hidden layer with no inter-layer connections.

So I’ve modified some Matlab code to allow the network to learn handwritten digits from the MNIST handwriting database. It’s pretty nifty and can reconstruct noisy images based on what it’s learned by itself (no labels involved).

Here’s a screenshot of a run using 100 digits from the database. The network has 300 hidden nodes. The images have 20% noise before reconstuction.


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s

%d bloggers like this: