Recent Developments in Deep Learning

Google Tech Talk
March 19, 2010

ABSTRACT

Presented by Geoff Hinton, University of Toronto.

Deep networks can be learned efficiently from unlabeled data. The layers of representation are learned one at a time using a simple learning module that has only one layer of latent variables. The values of the latent variables of one module form the data for training the next module. Although deep networks have been quite successful for tasks such as object recognition, information retrieval, and modeling motion capture data, the simple learning modules do not have multiplicative interactions which are very useful for some types of data.

The talk will show how to introduce multiplicative interactions into the basic learning module in a way that preserves the simple rules for learning and perceptual inference. The new module has a structure that is very similar to the simple cell/complex cell hierarchy that is found in visual cortex. The multiplicative interactions are useful for modeling images, image transformations, and different styles of human walking.

Speaker bio: http://www.cs.toronto.edu/~hinton/bio.html

Category:

Science & Technology

Tags:

• google
• tech
• talk
• machine
• learning

[coastwalker]

codifying learning is a tricky question it turns out, keep up the good work!

[Mark7BD]

oh gosh this brings back memories of university... though I was interested in the material.. I wish I spent more time applying this instead of sitting in a lecture.

[killgreed]

NERDS!!

[j27cn49s]

great talk

[Shizzmonger]

tldnw

[paucarre]

For those who have difficulties to understand this: the method is based on mapping inputs (images) to a sparse output which is easier to classify for a NN or a SVM. The improvement is based on adding a new mapping from images to the variance of those images. This way, there is more information so classifiers (NN and SVM) are better at predicting.

[FastFootElectro]

1:05:08 Monster