Take the 2-minute tour ×
Mathematics Stack Exchange is a question and answer site for people studying math at any level and professionals in related fields. It's 100% free, no registration required.

In my earlier question I asked about a technical aspect of solving a system of equations arising from looking for an entropy-maximizing distribution $p(x)$ continuous on $\mathbb{R}$ and constrained by KL-divergence with a zero-mean Gaussian distribution. That is, in addition to the usual probability density and variance constraints, I have the following constraint for $p(x)$:

$$D(p_N(y)\|p(y))=\int_{-\infty}^{\infty}\frac{1}{\sqrt{2\pi}\sigma_N}e^{-y^2/2\sigma^2}\log\frac{\frac{1}{\sqrt{2\pi}\sigma}e^{-y^2/2\sigma^2}}{p(y)}dy<\epsilon$$

Thanks to user anon, the form of the function $p(y)$ was found, but it is not a density function and now I am trying to interpret why is this the case.

First, here is the system of equations (copied from the earlier question) that I derived using Calculus of Variations (and help from Gallager's "Information Theory and Reliable Communication":

$$\begin{align} 0&=\log(p(y))+1-\lambda-\gamma y^2-\eta \left(\frac{e^{-y^2/2}}{\sqrt{2\pi}}\right)\left(\frac{1}{p(y)}\right)\\ 0&=1-\int_{-\infty}^{\infty}p(y)dy\\ 0&=1-\int_{-\infty}^{\infty}y^2p(y)dy\\ 0&=c+\int_{-\infty}^{\infty}\frac{e^{-y^2/2}}{\sqrt{2\pi}}\log(p(y))dy \end{align} $$

(for simplicity I set $\sigma=1$; $c=\epsilon+\frac{1}{2}\log(2\pi )$)

From anon's helpful comment, we can actually solve the first equation in terms of Lambert W function to obtain the following:

$$p(x)=\frac{\eta e^{-y^2/2}}{\sqrt{2\pi}W(e^{-(1+2\gamma)y^2/2+(1-\lambda)})}$$

When $|y|\rightarrow\infty$, $e^{-ay^2+b}\rightarrow 0$, and since $W(0)=0$, $p(y)\rightarrow\infty$. Thus, this is obviously not a pdf!

This is entirely due to the KL-divergence constraint (very similar situation arises when variance constraint is removed). How does one explain this? There are obviously probability distributions that meet the KL-divergence constraint (e.g. a Gaussian with appropriately picked variance). Does this mean that optimal distribution does not exist, and all distributions one can try would be sub-optimal? Is there a rigorous explanation for this?

Perhaps I did something wrong? Is there another method I should have employed?

share|improve this question
3  
Stupid question: is $1 + 2\gamma$ necessarily positive? –  Rahul Sep 22 '11 at 5:21

1 Answer 1

up vote 3 down vote accepted

You established the indeterminate form $0/0$; from this you cannot casually conclude $p\to\infty$. In the last line of my comment that you cite, I used the trick $e^{W(t)}=t/W(t)$ in order to simplify the expression - but if you want to look at it asymptotically, you can also analyze the prior form: $$p(y)=\exp(-\alpha+W(e^{\alpha}\beta)),$$ where $\alpha=1-\lambda-\gamma y^2$ and $\beta=\eta e^{-y^2/2}/\sqrt{2\pi}$. Now if $\gamma\in[-1/2,0)$, we get the form $p=e^{-\infty+0}$, so we find that $p\to0$. For $\gamma<-1/2$, you'll have to find a way to show $$W(ae^{-(1/2+\gamma)y^2})+\gamma y^2\to-\infty \text{ as } |y|\to\infty.$$ Above $a>0$ is arbitrary (and $\eta e^{1-\lambda}/\sqrt{2\pi}$ in particular). Note $W(x)>1$ for $x>e$, so we have $$e^W\le W e^W =x$$ hence $W\le\log x$ for sufficiently large $x$, which makes the above $-O(y^2)\to-\infty$ (allowing notational sloppiness), proving $p\to0$ as $|y|\to\infty$ whenever we have $\gamma<0$.


This only addresses your claim that $p$ doesn't vanish at the extremes. For your broader questions concerning maximum entropy, KL divergence, or the original optimization problem, I really have no idea.

share|improve this answer
    
Full disclosure: I've no love for the tendency to "simplify" $\exp(W(t))$ into $t/W(t)$ myself; the latter has a completely unnecessary removable discontinuity among other things... –  J. M. Sep 22 '11 at 8:02
    
@J.M.: Hm, I suppose you're right. I don't work with $W$ often so when I originally commented it seemed like the form with no function composition would be "simpler." –  anon Sep 22 '11 at 8:16
    
It's just my opinion, of course. Maybe there's a situation when $t/W(t)$ is a better form, but I haven't seen one. –  J. M. Sep 22 '11 at 8:19
    
Sorry, I had a family emergency of sorts and haven't been able to reply here. First of all, thank you again, anon. I am appreciative of your time -- you have really elucidated this for me. –  M.B.M. Sep 27 '11 at 4:26

Your Answer

 
discard

By posting your answer, you agree to the privacy policy and terms of service.

Not the answer you're looking for? Browse other questions tagged or ask your own question.