prediction
esiegel | April 18, 2008Bret Underwood, a friend of mine from my time in Madison, WI, saw my post on String Theory, and took issue with my statement that it wasn't testable. I'm still standing behind what I said, but let's address what Bret has to say.
I donât understand your argument above for why string theory is untestable. In fact, it seems to me you just outlined the best possible case for string theory! What you said above is that if I have a string theory construction of a phenomenon (say, the Standard Model or Inflation), which uses a set of parameters X, and makes some predictions, then I can find another…
developingintelligence | November 19, 2007Geoff Hinton has a new TiCS paper describing recent advances in algorithms used to train multilayered neural networks.
First, a little background: neural networks of a sufficient size can calculate any mathematical function (an infamous proof among neural network modelers). Unfortunately, the tricky part is figuring out how to set the connections in that network to calculate those functions.
This is where learning algorithms become necessary - unless you want to tweak each connection by hand until you get a working network (not a problem if you don't care how the brain works), then you…
illconsidered | March 26, 2006This is just one of dozens of responses to common climate change denial arguments, which can all be found at How to Talk to a Climate Sceptic.
Objection:
Climate is an inherently chaotic system, and as such it can not be predicted.
Answer:
Firstly, let's make sure we define climate. Climate is generally viewed as an average of weather patterns over some meaningful time period. The number of years may vary and there are probably plenty of other finer points to quibble about in there, but the purpose of getting this definition out in front is to be sure we are safe in discounting the very…