Abstract: I'll describe how I came to two very different results related to machine learning. One of those results is a uniqueness theorem: Under generic conditions, two neural nets that do exactly the same thing have the same architecture and parameters. The other result pertains to the problem of fitting a smooth (C^m) function to data.

Click here to convert to your local time: 27th April 15:00 CET

Watch the talk again:

Read or download the slides: