[ml] Machine Learning Lecture Wed 5/5: A Brief Tour of Statistics
Thomas Lotze
thomas.lotze at gmail.com
Thu Apr 29 04:28:07 UTC 2010
Hey all! Next week, I'm going to try to give a brief tour of statistics:
how to think like a statistician, what statisticians are trying to do, and
some big statistical ideas relevant to machine learning. I'm hoping to give
people a good understanding of statistical thinking, and hopefully spawn
some ideas on future talks. We'll meet at 8 PM in the same place we've been
meeting (the classroom in the back corner)
I'm planning on starting fairly promptly (8:05) regardless of how many
people are here, because there's a lot I'm hoping to talk about -- so please
arrive as close to 8 PM as possible!
The topics I'm planning on covering are:
- Basic statistics and probability: random numbers and distributions
- Some good distributions to know: binomial, poisson, normal
- Estimating distribution parameters
- Unbiased estimators and the bias/variance tradeoff
- Confidence intervals, Testing and p-values
- Maximum Likelihood Estimators
- Another look at Regression
- Logistic Regression
- Why statisticians love the normal distribution
- Why Taleb Naseem hates the normal distribution
- Bootstrapping techniques
- A statistical perspective on Induction
- What Bayesian Statistics is basically about
But if you have anything else you're interested in, let me know!
Cheers,
Thomas
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://www.noisebridge.net/pipermail/ml/attachments/20100428/110f4d57/attachment.html>
More information about the ml
mailing list