4

Let $z \in \mathbb{C}$ and let $W$ be the Lambert $W$ function. In this post it is shown that if $|W(-\ln z)| > 1$ then the infinite power tower $z^{z^{z^{z^...}}}$ does not converge, that is $|W(-\ln z)| \leq 1$ is a necessary condition for the convergence of $z^{z^{z^{z^...}}}$.

Here I would like to show that $|W(-\ln z)| < 1$ is a sufficient condition, that is if $|W(-\ln z)| < 1$ then $z^{z^{z^{z^...}}}$ is convergent.

cpiegore
  • 1,490
  • https://en.wikipedia.org/wiki/Lambert_W_function#Example_3 Seems to have something that might be of interest. – Simply Beautiful Art Mar 13 '16 at 22:27
  • I already knew about that. It's worth pointing out that the formula actually gives a result for every complex number except $0$ and $1$ – cpiegore Mar 13 '16 at 23:39
  • 2
    Oh, that's good! Now you change the condition of sufficiency from $\le 1$ to $\lt 1$. Keep up the good work! –  Mar 14 '16 at 00:47
  • @YiannisGalidakis I only did that because I cannot prove the sequence is convergent when $|W(-\ln z)| = 1$. By the same token, nobody has given a convincing argument for why the sequence should diverge, at least not in my opinion. – cpiegore Mar 14 '16 at 01:03
  • 4
    You should have asked a different question anyway. Changing the question after it has received two correct answers, invalidates both these answers, since we are answering a different question. As far as the answers, a sequence can be many other things, not limited to "convergent/divergent". It can also be periodic, with ANY period $p>0$ and it can also be "chaotic" (or "wandering"- something which is happening here). For wondering sequences, the only proof that they are such depends on proving that they are randomly dispersed. Such sequences are neither "diverging" nor "converging". –  Mar 14 '16 at 01:22
  • @YiannisGalidakis Can you explain what you mean by "divergent." To me a sequence that "diverges to $\infty$", periodic sequences and "chaotic/wandering" sequences are all examples of divergent sequences. In other words divergent simply means $\lim_{n \to \infty}a_n$ does not exist. – cpiegore Mar 14 '16 at 01:31
  • @cpiegore : why didn't you ask about $c_{n+1} = b(e^{c_n}-1)$ as I did in the other post ? in my opinion it is a easier to visualize sequence – reuns Mar 14 '16 at 01:43
  • I don't think you'll arrive a an *own* proof; I've read the article of D.Shell (it is online somewhere, look perhaps in the database of the tetration-forum for more searchable information or a link) and it is difficult - for me ... well, I just accepted that there is a proof ;-) . Unfortunately the Euler-proof which is easier reflects only real bases b and u and so is not sufficient for you. – Gottfried Helms Mar 14 '16 at 01:43
  • You can use the term "divergent" as a catch-all for "non-convergent" sequences (including periodic ones), but not with "wandering" ones. It's not a question of disagreement in the definition, it's because you don't know what they do (or what they will do next). They may decide to converge after the 1000000000-th term or go to $\infty$, for example). Precisely because you don't know (unless you calculate the next terms), you cannot characterize them as either. –  Mar 14 '16 at 01:43
  • I think the strict dichotomy between "convergent" and "divergent= not convergent" stems from real analysis and I know this strict dichotomy for instance in the discussion of summation-procedure of series. A very simple example that this strict dichotomy is uneasy for complex analysis gives an example from plane geometry, namely the spiral. Here it is much intuitive to introduce a *trichotomy* between spirals which contract, which expand and which have a constant radius, thus are circles. The term "wandering" seems also meaningful, see for analogy the perihel-movement of the planet Mercur. – Gottfried Helms Mar 14 '16 at 01:51
  • @GottfriedHelms What do you mean by own? I'm guessing it means if-and-only-if. – cpiegore Mar 14 '16 at 01:55
  • @cpiegore: I mean with "own": "on your own" - to do the proof "on your own". For me this were too difficult, I thought the same will be true for you (but of course may be this is actually not more than the level of some homework at the college - don't know, haven't studied math as main subject) – Gottfried Helms Mar 14 '16 at 01:58
  • @YiannisGalidakis What is your problem? I am not trying to be rude or fraudulent. I changed the post because I am VERY confused in the case $|W(-\ln z)| = 1.$ – cpiegore Mar 14 '16 at 02:35
  • @cpiegore it is merely frowned upon to edit the question drastically when people have already put hard work into answers for you. It is recommended you start anew with a new post if this is the case, and post a link so anyone wanting to answer your other questions may find them. – Simply Beautiful Art Jun 30 '16 at 16:34
  • I don't think the power tower has any problems with convergence at $z=1$, it is fairly obvious that $1^{1^{1^{\dots}}}=1$, as for using the formula $z^{z^{z^{\dots}}}=\frac{W(-\ln(z))}{-\ln(z)}$, one could instead use $e^{-W(-\ln(z))}$ to arrive at the conclusion for $z=1$, it still equals $1$, or one could use limits and L'Hospital's rule. – Simply Beautiful Art Jun 30 '16 at 16:38
  • @Simple Art I wasn't talking about the point $z = 1$. I was talking about an entire class of values, namely $\lbrace z \in \mathbb{C} : |W(-\ln z)| = 1 \rbrace$ (the boundary of the Shell-Thron Region) where W is the Lambert W Function. My question has since been answered in detail in a subsequent post. – cpiegore Jun 30 '16 at 17:48

3 Answers3

8

The key concept is here the "Shell-Thron-region". In articles in the previous century initially W. Thron and later D. Shell based on Thron's work proved that if you have a complex base, say $b$ such that $b=t^{1/t}$ or, with $u=\log(t)$, such that $b=\exp(u \exp(-u))$ then the infinite power tower converges if $|u| \lt 1$ and the point of convergence is $t$. (See my earlier picture in MSE where I've related those 3 variables with each other)

The numerical values given in Yiannis Galidakis' answer have $|u|=1-\varepsilon$ so the iteration should converge although very slowly. I found, that a nice picture occurs if you separate the trajectory in 4, or even better: 72 subtrajectories. With Pari/GP and 800 digits precision you get a nice shape which has some "fractal-like" or "snowflake-like" border. I've done the iterations from $z_0=1$ to up to 80 x 72 iterations so each partial curve has 80 points, nearly neighboured with each other - and each pair of neighboured points of the same color has distance of 72 iterations; for a real good image one should proceed to at least 72^3 x 72 points to get a valid impression that this strangely shaped curve really contracts. See a q&d picture made with Excel using values made with Pari/GP, 800 digits precision:

q&d picture

One recognizes 4 segments which together do about one round. These are the first four segments of 72 segments, (where the 5 'th would nearly overlap the first, the 6 'th nearly the second and so on, but are not shown here to keep the picture clean). The brown segment is the 32 'th and its small additional excess from the blue one shows that the expected contraction is -at least- not smooth.
I've no nerve to increase the number of points at the moment (it's night here), possibly my hints give enough ideas to proceed on your own.

[update] I couldn't stop to try to discern the convergence. It appears, that not only in steps of 72 the iterates are tight together, but that it needs 322 of such 72-steps to fill one round of the curve. So I took an arbitrary initial value from my existing list, $\small y_0=-0.5602531521 - 0.6868631844 I$, then iterated 322*72=23184 times to arrive at $ \small y_{23184} \approx -0.5602563718 - 0.6868510240 I$ and proceeded 20 times with that iteration width. The following protocol show contraction but only in the fourth decimal digit of the distance to the fixpoint t, of course not visible in the picture:

    real y_k       imag y_k       | y_k - t| =distance to fixpoint 
  -0.5602563718  -0.6868510240  0.8863698615
  -0.5602654611  -0.6868245642  0.8863551032
  -0.5602477332  -0.6867936307  0.8863199274
  -0.5602391855  -0.6867763400  0.8863011262
  -0.5602265922  -0.6867593629  0.8862800106
  -0.5602215274  -0.6867495245  0.8862691855
  -0.5602178010  -0.6867358203  0.8862562109
  -0.5602148750  -0.6867278280  0.8862481684
  -0.5602175553  -0.6867173579  0.8862417497
  -0.5602183774  -0.6867059453  0.8862334262
  -0.5602266751  -0.6866972799  0.8862319570
  -0.5602460790  -0.6866801357  0.8862309393
  -0.5602492634  -0.6866452713  0.8862059387
  -0.5602499101  -0.6866233340  0.8861893503
  -0.5602465541  -0.6865997931  0.8861689891
  -0.5602452598  -0.6865858545  0.8861573713
  -0.5602473046  -0.6865693793  0.8861458993
  -0.5602478202  -0.6865574553  0.8861369869
  -0.5602548561  -0.6865457844  0.8861323930
  -0.5602616629  -0.6865325382  0.8861264340

[end update]

That is different with values b where the according value of u is $|u|=1$ and thus lie on the boundary of the complex unit disk. An example has been given in the comment at Yiannis Galidakis' answer with u as some complex unit root. Then we have no convergence and the curve (with roughly same shape as the beginning of the shown curve) does not contract but has its trajectory "stationary" - I called this, when I've seen it first time, "equator" because it reminds me to the meridians on a globus - not disappearing away from and also not contracting towards the fixpoint, but of course there are established technical terms for such -thanks to Yiannis pointing me to this in a comment last night.


P.s.: to improve numerical stability and computing speed when many iterations are needed, use the following conjugacy relation:
the original iteration demands
  • use some $\small z_0$,
    compute $\small z_{k+1} = b^{z_k}$ and iterate
    until some $\small z_n$ .

You can do the following replacement:

  • use the same $\small z_0$,
    compute $\small y_0=(z_0/t)-1$ ,
    compute $\small y_{k+1}=\exp(u \cdot y_k) - 1$ for as many iterations as before
    to get $\small y_n$,
    then compute $\small z_n = (y_n+1) \cdot t$
    $ \qquad \qquad $ with $\small t=\exp(-W(-\ln(b)))$ and $\small u=\ln(t)=-W(-\ln(b))$
4

Here I would like to show that $|W(−\ln(z))|\le 1$ is also a sufficient condition, that is if $|W(−\ln(z))|\le 1$ then $z^{z^{z^{\ldots}}}$ is convergent.

It's not true. Take $c=2.043759690+0.9345225945i$. Then (with some Maple code:)

restart;
with(plots);
F := proc (z, n)#power tower recursively defined
option remember;
if n = 1 then z else z^F(z, n-1)
end if
end proc;
W := LambertW;
c := 2.043759690+.9345225945*I;
evalf(abs(W(-ln(c))));

0.99999999

L := [seq(evalf(F(c, n)), n = 1 .. 100)];
complexplot(L, style = point);

Here's the list of values $c,c^c,c^{c^c},\ldots$ plotted against the Complex plane:

enter image description here

  • I don't think your value $c$ is a counterexample. According to my calculator a TI-89 titanium $|W(-\ln c)| = 0.99999999994964$. – cpiegore Mar 13 '16 at 20:50
  • 1
    Just for good measure, I repeated the calculation on an arbitrary precision calculator program. I have computed that, correct to 40 digits $W(-\ln c = -0.0874257267 7360635585 5724509847 8820858887 8 -0.9961710406 3382091178 7513030479 3704246519 i$ and $|W (-\ln c)| = 0.9999999999 4968147266 6848075038 9069157318$ – cpiegore Mar 13 '16 at 20:58
  • That's because I gave $c$ in numerical form. Try it with $c= \exp(\exp(2i\pi\sqrt{5})/\exp(\exp(2i\pi\sqrt{5})))$. Make sure you scan the text properly. –  Mar 13 '16 at 21:11
  • You're still wrong: using the same program I have computed that $|w(-\ln(c)| = 1+9.7463478*10^{-76}$ – cpiegore Mar 13 '16 at 21:27
  • 1
    Disregard my last comment. Upon repeating the calculations with higher precision I obtained $|W(-\ln c)| - 1$ is on the order of $10^{-1000}$. Nonetheless you have NOT proved that the sequence is divergent. – cpiegore Mar 13 '16 at 22:05
  • Of course I have NOT proved that the sequence is "divergent". Because it isn't. I have only provided a numerical counterexample to your claim that the condition is sufficient. In other words, that the sequence may not (necessarily) be convergent. Here, "not convergent" isn't the same as "divergent". –  Mar 13 '16 at 22:39
  • I find for your $c$, $c^{c^{c^{\unicode{x22F0}}}}$ converges to $0.593177 \dots + 0.916084 \dots \mathrm{i}$. – Eric Towers Mar 13 '16 at 22:51
  • @ Eric Towers HOW – cpiegore Mar 13 '16 at 23:20
  • @YiannisGalidakis Your last comment makes absolutely no sense. In my mind a sequence that does not converge is by definition divergent. – cpiegore Mar 13 '16 at 23:23
  • How do EITHER of you know that the sequence is "not divergent" ie it is convergent. Keywords: HOW ; DETAILS – cpiegore Mar 13 '16 at 23:28
  • @YiannisGalidakis : I don't know what you proved or not, but are you sure your "non-convergence" doesn't come from numerical imprecision ? – reuns Mar 14 '16 at 01:41
  • 4
    It's precisely the opposite of what you think: An apparent "convergence" will come from numerical imprecision (like Gottfried says, which you apparently haven't seen yet). Such are examples where there's always imprecision because they come from irrational multiples of the phase angle on the unit circle, so they are never represented correctly, even when you use 10 million digits. They are examples of chaotic sequences. If you reduce accuracy to 2-3 digits, they may converge. –  Mar 14 '16 at 01:49
  • @YiannisGalidakis Interesting concept of "wandering sequences". When googled, I find some Ergodic theory, which sounds like it is probably fairly recent. I'm 'wandering' if there is a formal way to proving a sequence is wandering or not. – Simply Beautiful Art Jun 30 '16 at 16:44
  • @SimpleArt I am not certain about it, but I doubt it. Questions like this clearly go beyond the scope of plain deterministic dynamics, since the ergodicity of these cases seems to lie half-way between determinism and chaos. And I am not sure what kind of definitions one can use for this. Cheers. –  Jun 30 '16 at 17:14
2

Here I want to show another argument, why convergence occurs with the type of iteration as in the title.

Only let me rename the involved variables to my own year-long use for my own comfort when writing this:

  • $b$: I use $b$ for "(b)ase" in $z_{k+1}=b^{z_k}$ with complex numbers $z_k$

  • $t$: Then I use $t$ for the "fixpoint" such that $t=\lim_{k \to \infty} z_k$
    $\qquad $(if this is convergent, otherwise if the inverse iteration $z_{k+1} = \log_b(z_k)$ converges to the fixpoint or at least if the Newton-iteration gives such a fixpoint)

  • $u$: For the log of the fixpoint I use $u$ such that
    $ \qquad b= t^{1/t} = \exp(u \cdot \exp(-u))$ or
    $ \qquad t=\exp(-W(-\log(b)))$ and
    $ \qquad u=-W(-\log(b))$

  • conjugacy: Instead of writing $z_{k+1}=b^{z_k}$ it is equivalent to write $y_{k+1}=t^{y_k}-1$ and use conjugacy
    $\qquad y_k = z_k/t-1$ and $z_k = (y_k+1)\cdot t$.
    $\qquad$ Apart of the advantages which I exploit below, it seems from some heuristics that this might also be numerically more stable than the iterated computation of the un-conjugated original function.


For the conjugated version we can generate the Schroeder-scheme (see for instance wikipedia) for the implementation of the iteration (which, if $t$ is real and $t \in (e^{-1},e) $ can even be fractionally iterated) . With introduction of the Schroeder-function $\sigma()$ and its inverse $\sigma^{-1}()$ this means to compute for some iteration "height" $h$: $$ y_{k+h} = \sigma^{-1} (u^h \cdot \sigma (y_k)) \tag 1$$
The original trajectory $\{z_0,z_1,z_2,\cdots \}=\{1,b,b^b, \cdots \}$ (possibly converging to $t$) conjugates to $\{y_0,y_1,y_2,\cdots \}=\{1/t-1,b/t-1, \cdots \}$ (possibly converging to $0$ which is $0 =t/t-1$ by the conjugacy) . Inserting $y_0$ in the Schroeder-equation (1) we have $$ y_{h} = \sigma^{-1} (u^h \cdot \sigma (y_0) \tag 2$$
The iteration "height" $h$ occurs here only in the exponent of $u$.
The Schroeder-function $\sigma()$ has an invertible formal power-series for $|u| \ne 1$ with constant term $s_0=0$.

Now, if $|u| \lt 1$ then for $h$ going towards positive infinity the cofactor $u^h$ decreases to zero and the whole expression gives $$ \sigma^{-1} ( 0 ) = 0 \tag 3$$ and conjugacy gives $ (0+1)\cdot t = t$ which is the fixpoint $t$ of the original iteration.

So this should be sufficient for another proof of convergence (according to the OP's title) "on one's own" .


Problems occur, if $|u|=1$.

In the denominators of the coefficients $s_k$ of the Schroeder-function $\sigma(y) = s_1 y/1! + s_2 y^2/2! + s_3 y^3/3! + ... + s_k y^k/k! + ... $ we have products of $(u^i-1)$ for $i=1..k$, so if $u=1$ or some $u^i =1$ all following coefficients become singular and the Schroeder-function $\sigma()$ does not exist for such $u$ where for some $k \in \mathbb N^+$ we have $u^k=1$. (see for more explanation of this my essay functional iteration (pdf) explicitely for the details of the denominators pg 25/26 or more concisely the other essay Eigendecomposition (html) )

However, when $|u|=1$ and $u=\exp(2 \pi î /c)$ with some irrational $c$ the denominators of the coefficients of $\sigma()$ don't show that singularities and further analyses based on the Schroeder-mechanism with the conjugation to $y_k$ might thus be possible this way.

Remark: unfortunately this nonexistence of a $\sigma()$ for $u=\exp(2 \pi î /c)$ with $c$ rational does not allow to analyze by this ansatz the reason, why then also convergence occurs. In case I'll find some operational idea for this I'll add this here, although the OP's question is not at this point.

  • I'd just like to say that I would I greatly appreciate any insight you could give for the case $|u |= 1$ – cpiegore Sep 27 '18 at 00:44
  • 1
    I have seen Baker and Rippon's proof of convergence for rational $c$, but I find it to be complex and hard to grasp. – cpiegore Sep 27 '18 at 00:48
  • @cpiegore - yes, I'd like to find some real proof , or path to a proof, too. All I have so far are that observations/numerical evidence : 1) of contraction of the orbit if $c$ is a rational, 2) and of the (single) slingshot towards (complex) infinity ($\log_b(0)$) in a seemingly non-contracting orbit on a fractal border becoming more and more dense when the steps of iterations follow the convergents of the continued fraction of the irrational $c$. For Liouville-like $c$ parts of the orbit simulate the visual image of some rational $c$ and "pretend" to be so. (I'm still generating pictures;)) – Gottfried Helms Sep 27 '18 at 07:08
  • (...) did you see my "equator"-text? I've already linked at it in some other comments of one of your questions, I think. (http://go.helms-net.de/math/tetdocs/_equator/mse160614_equator.pdf ). Again: only observations so far... – Gottfried Helms Sep 27 '18 at 07:10