Recent Developments in Deep Learning
Geoff Hinton, 30 May 2013
www.youtube.com/watch?v=vShMxxqtDDs (65 minutes)
Neural nets have been around for a long time, but people could not get them to work as well as they had hoped. The explanation for this was that learning by backpropagation would never give better results because:
- It required labelled training data.
- Learning time was very slow with deep nets.
- It got stuck in local optima.
Monads and Gonads
Douglas Crockford, 15 January 2013
http://www.youtube.com/watch?v=b0EF0VTs9Dc (49 minutes)
Mission Impossible: Constructing Charles Babbage's Analytical Engine
Doron Swade, 8 May 2012
http://www.youtube.com/watch?v=FFUuN-ZRLz8 (87 minutes)
This is longer than the others, but there's really no heavy-lifting in this talk. Swade outlines the very ambitious project to complete the design and construction of Charles Babbage's Analytical Engine. (If you want to support the effort, you can make a contribution at plan28.org.)