Jump to:  Browse | Search | Email 

Browse:  most popular entries | oldest entries | most recent entries 

Information Theory, PR and NN

I have been told that information theory disproves the theory of evolution. Is that true? If so, what book or source material is available on the subject?

What are the prerequisites for the information theory course?

Is Typicality on the syllabus?

I can't distinguish the line styles in fig 3.2, the probability density P(x|l) as a function of l.

The negative part of Shannon's noisy coding theorem contains a function H2^-1(1- R/C) what does " H sub 2 to the minus 1" mean? Is it the inverse function, ie. the function such that ((H sub 2 to the minus 1) acting on (H sub 2) acting on x) gives 1?

I have a problem with how to calculate the bit error of the (7,4) Hamming code. Specifically, I can't follow the assertion "when the noise vector has weight two, the parity bits and source bits are equally likely to be flipped, so the bit error is 3/7 the block error"

1. Regarding the inference of sigma_n and sigma_n-1 from data, for Gaussians. This sounds suspiciously like the maximum likelihood value for sigma is changing depending on whether we also calculate xbar or not. Is this true, or am I confused? 2. What is the difference between max likelihood and max posterior probability? Is it just the inclusion of a non-uniform prior?

p380. top paragraph: var(PHI^) = var(phi)/R where PHI^ is our estimate of mean of phi from taking R samples obeying a probablity distribution, P. How do you prove this?

Showing 1 to 8 of 8 entries