Informations

1 Connection to statistics

See Shalizi, Note: Information Theory, Axiomatic Foundations, Connections to Statistics.

Many connections, not always clear because the continuous probability spaces of statistics are far from the discrete spaces of coding theory. Nonetheless, there are contacts, most glaringly via the K-L divergence.

2 Shannon Information

“I have a discrete random process, how many bits of information do I need to reconstruct it?”

Vanilla information, thanks be to Claude Shannon. A thing related to coding of random processes.

Given a random variable \(X\) taking values \(x \in \mathcal{X}\) from some discrete alphabet \(\mathcal{X}\), with probability mass function \(p(x)\).

\[ \begin{array}{ccc} H(X) & := & -\sum_{x \in \mathcal{X}} p(x) \log p(x) \\ & \equiv & E( \log 1/p(x) ) \end{array} \]

More generally if we have a measure \(P\) over some Borel space \[ H(X)=-\int _{X}\log {\frac {\mathrm {d} P}{\mathrm {d} \mu }}\,dP \]

Over at the functional equations page is a link to Tom Leinster’s clever proof of the optimality of Shannon information via functional equations.

One interesting aspect of the proof is where the difficulty lies. Let \(I:\Delta_n \to \mathbb{R}^+\) be continuous functions satisfying the chain rule; we have to show that \(I\) is proportional to \(H\). All the effort and ingenuity goes into showing that \(I\) is proportional to \(H\) when restricted to the uniform distributions. In other words, the hard part is to show that there exists a constant \(c\) such that

\[ I(1/n, \ldots, 1/n) = c H(1/n, \ldots, 1/n) \]

for all \(n \geq 1\).

Venkatesan Guruswami, Atri Rudra and Madhu Sudan, Essential Coding Theory.

3 K-L divergence

Because “Kullback-Leibler divergence” is a lot of syllables for something you use so often. Or you could call it the “relative entropy”, if you want to sound fancy and/or mysterious.

KL divergence is defined between the probability mass functions of two discrete random variables, \(X,Y\) over the same space, where those probability mass functions are given \(p(x)\) and \(q(x)\) respectively.

\[ \begin{array}{cccc} D(P \parallel Q) & := & -\sum_{x \in \mathcal{X}} p(x) \log p(x) \frac{p(x)}{q(x)} \\ & \equiv & E \log p(x) \frac{p(x)}{q(x)} \end{array} \]

More generally, if the random variables have laws, respectively \(P\) and \(Q\): \[ {\displaystyle D_{\operatorname {KL} }(P\|Q)=\int _{\operatorname {supp} P}{\frac {\mathrm {d} P}{\mathrm {d} Q}}\log {\frac {\mathrm {d} P}{\mathrm {d} Q}}\,dQ=\int _{\operatorname {supp} P}\log {\frac {\mathrm {d} P}{\mathrm {d} Q}}\,dP,} \]

4 Jenson-Shannon divergence

Symmetrised version. Have never used.

5 Mutual information

The “informativeness” of one variable given another… Most simply, the K-L divergence between the product distribution and the joint distribution of two random variables. (That is, it vanishes if the two variables are independent).

Now, take \(X\) and \(Y\) with joint probability mass distribution \(p_{XY}(x,y)\) and, for clarity, marginal distributions \(p_X\) and \(p_Y\).

Then the mutual information \(I\) is given

\[ I(X; Y) = H(X) - H(X|Y) \]

Estimating this one has been giving me grief lately, so I’ll be happy when I get to this section and solve it forever. See nonparametric mutual information.

Getting an intuition of what this measure does is handy, so I’ll expound some equivalent definitions that emphasise different characteristics:

\[ \begin{array}{cccc} I(X; Y) & := & \sum_{x \in \mathcal{X}} \sum_{y \in \mathcal{Y}} p_{XY}(x, y) \log p(x, y) \frac{p_{XY}(x,y)}{p_X(x)p_Y(y)} \\ & = & D( p_{XY} \parallel p_X p_Y) \\ & = & E \log \frac{p_{XY}(x,y)}{p_X(x)p_Y(y)} \end{array} \]

More usually we want the Conditional Mutual information.

\[I(X;Y|Z)=\int _{\mathcal {Z}}D_{\mathrm {KL} }(P_{(X,Y)|Z}\|P_{X|Z}\otimes P_{Y|Z})dP_{Z}\]

See Christopher Olah’s excellent visual explanation.

6 Kolmogorov-Sinai entropy

Schreiber says:

If \(I\) is obtained by coarse graining a continuous system \(X\) at resolution \(\epsilon\), the entropy \(HX(\epsilon)\) and entropy rate \(hX(\epsilon)\) will depend on the partitioning and in general diverge like \(\log(\epsilon)\) when \(\epsilon \to 0\). However, for the special case of a deterministic dynamical system, \(\lim_{\epsilon\to 0} hX (\epsilon) = hKS\) may exist and is then called the Kolmogorov-Sinai entropy. (For non-Markov systems, also the limit \(k \to \infty\) needs to be taken.)

That is, it is a special case of the entropy rate for a dynamical system — Cue connection to algorithmic complexity. Also metric entropy?

7 Relatives

8 Estimating information

Wait, you don’t know the exact parameters of your generating process a priori? You need to estimate it from data.

9 Connection to statistical mechanics

See Entropy vs information.

10 References