We will learn to model problems and systems using mathematics and computers. We'll be using the python computer language (which I will teach everybody at the start of the course). We'll cover statistical models, automata models, and classical applied-math models.
We'll also discuss the nature of modelling, based on readings from Nate Silver's Signal and the Noise.
At the end of the course, you'll have the start of a very valuable skill set.
Canopy - python software for scientific computing that we'll make use of in class.
We were discussing Chapter 8 in class on Bayesianists vs. Frequentists. Prof. Ipeirotis has a nice post giving a better explanation about the difference between the frequentist and Baysian perspectives. Prof. Wasserman, from Carnegie Mellon, has a blog post, courtessy of Corrine, that argues Nate has been mis-informed about the difference between Baysians and Frequentists, and that he's actually a frequentist. Of course, this started an argument, which get's down into the weeds and doesn't really settle anything unless you want to dig in deeply. To balance the picture, here's a different blog post from a recent convert to the frequentist philosophy.
Doug point's out that there is a lecture of interest to our climate modelling section this coming Friday, April 4, 2014 2:30 PM, 113 Information Sciences and Technology Building, The Cybertorium
"Big data" is a dominate idea in technological development right now, closely related to our readings of Nate Silver's book. But there is also a lot of pushback right now. Poster-child of model-free big-data inference Google flu trends seemed to work pretty good at first. Now there's another effort-involving critical analysis of this by Olson, Konty, Paladini, Viboud, and Simonsen as well as cheap commentaries by Butler, Lazer, Kennedy, King, and Vespignani, and Tim Harford helping the limitations of data without models.
Stephen Wolfram's enumeration of 1-dimensional spatial rewrite rules.