A family of level-transitive groups with positive fixed-point proportion and positive Hausdorff dimension

Imagine a vast, infinite forest where every tree branch splits into exactly the same number of smaller branches, forever. In the world of mathematics, this is called a "regular tree," and the creatures that live there are groups of "automorphisms"—think of them as magical dancers who can shuffle the branches around without ever breaking the tree's structure. Mathematicians love studying these dancers because their moves often hide deep secrets about numbers and equations. One of the biggest puzzles in this forest is figuring out how many of these dancers ever stop moving at all. Specifically, mathematicians ask: if you pick a dancer at random, what are the odds they will stand still on at least one path that goes all the way to infinity? This chance is called the "fixed-point proportion." For a long time, most experts believed that for most interesting groups of dancers, this chance was zero—meaning almost no one ever stops. But a few rare exceptions existed, and they were so complicated that nobody could calculate the exact odds for them.

This paper, written by Santiago Radi, introduces a new, clever way to build families of these mathematical dancers and, for the first time, calculates exactly how many of them stop moving. The author doesn't just find a few new examples; they create a whole factory for them. By using a specific recipe involving "iterated wreath products" (a fancy way of stacking layers of shuffling rules), Radi constructs groups that are not only level-transitive (they can reach any part of the tree) and have a "positive Hausdorff dimension" (they are large and complex enough to fill a significant chunk of the forest), but they also have a guaranteed, non-zero chance of stopping. Even better, the paper provides a clear formula to calculate this exact chance for any group built this way. This is a big deal because it solves a long-standing guess that such groups couldn't exist, and it gives mathematicians a precise tool to measure the "stopping power" of these complex structures, including those that appear in the study of polynomial equations.

The Dance of the Infinite Tree

Imagine an infinite tree where every branch splits into $d$ new branches, forever. Now, imagine a group of dancers who can rearrange these branches. They must follow strict rules: if they move a branch, they must move the whole subtree attached to it in a consistent way. Mathematicians call these dancers "automorphisms," and the group they form is a "group acting on a tree."

The big question this paper tackles is: How many of these dancers ever stop moving?

To understand this, picture a specific path going up the tree, from the bottom to the very top (an "infinite path"). A dancer "fixes" a path if, after they do their shuffle, that specific path looks exactly the same as before. The "fixed-point proportion" is simply the percentage of dancers in the group who fix at least one of these infinite paths.

For decades, mathematicians thought that for most interesting groups, this percentage was zero. It was like believing that in a massive, chaotic dance party, no one ever stands still. While there were a few known exceptions, they were rare, and calculating the exact percentage for them was nearly impossible. Some of these exceptions were related to "iterated Galois groups," which are groups that appear when you study the solutions to polynomial equations (like $x^d + 1 = 0$) over and over again.

The New Factory for Dancers

Santiago Radi's paper does two main things. First, it builds a new, general method to calculate the fixed-point proportion for a huge class of these groups. Second, it uses this method to construct a brand-new family of groups that are "level-transitive" (they can reach any level of the tree), have "positive Hausdorff dimension" (they are large and complex, not just tiny specks), and—most importantly—have a positive fixed-point proportion.

The author constructs these groups using a recipe involving two subgroups, $Q$ and $P$, which act like the rules for the dance. The groups are denoted as $G^P_Q$. The paper proves that if you choose these rules correctly, the resulting group will definitely have dancers who stop moving.

The Magic Formula

The paper's first major breakthrough is a formula to calculate the fixed-point proportion for "iterated wreath products." Think of a wreath product as a stack of layers. In the bottom layer, you have a set of allowed moves (permutations). In the next layer, you apply those same moves to every branch, and so on.

Radi defines a special polynomial, $f_S(x)$, based on the set of allowed moves $S$. The fixed-point proportion is simply the largest number between 0 and 1 that solves the equation $f_S(x) = x$.

• If the moves in $S$ are very chaotic (transitive), the answer is 0.
• If the moves are very restrictive (every move fixes at least one spot), the answer is 1.
• If the moves are in between, the answer is a specific number between 0 and 1, which can be calculated exactly.

This is a huge improvement over previous work, where such calculations were often impossible or only known to be zero.

The New Family of Groups

Using this formula, Radi constructs a specific family of groups, $G^P_Q$, that act on a tree with $d$ branches (where $d \ge 3$ and $d$ is not 2 mod 4). These groups are:

1. Self-similar: They look the same no matter how deep you go into the tree.
2. Level-transitive: They can move any branch at any level to any other branch at that level.
3. Large: They have a positive Hausdorff dimension, meaning they are "big" in a mathematical sense.
4. Stopping: They have a positive fixed-point proportion.

The paper explicitly calculates this proportion for two specific examples.

Example 1: The Modular Dance
The author constructs a group based on the integers modulo $d$. The fixed-point proportion turns out to be:
$$\frac{\#\{a \in (\mathbb{Z}/d\mathbb{Z})^\times : a - 1 \in (\mathbb{Z}/d\mathbb{Z})^\times\}}{\Phi(d)}$$
This simplifies to a product involving prime factors of $d$:
$$\prod_{p|d} \frac{p-2}{p-1}$$
This formula works perfectly when $d$ is odd. If $d$ is even, the proportion is zero.

Example 2: The Polynomial Connection
The most exciting part is that this construction isn't just abstract math. The paper shows that one of these groups is exactly the "iterated Galois group" of the polynomial $f(x) = x^d + 1$.
This means that for the polynomial $x^d + 1$, the group of symmetries of its infinite sequence of roots has a calculable, positive fixed-point proportion. The paper proves that for $d \ge 3$ (and $d$ not 2 mod 4), this proportion is:
$$\prod_{p|d} \frac{p-2}{p-1}$$
This is a concrete, explicit number that mathematicians can now use, whereas before it was a mystery.

What About the "Impossible" Cases?

The paper also addresses a specific case where $d \equiv 2 \pmod 4$ (like $d=2, 6, 10, \dots$). The author tried to build these groups for these values but found a "blockage." Using computer software (GAP), they checked all possible transitive groups for $d$ up to 30 and found that no group met the necessary conditions to have a positive fixed-point proportion. This suggests that for these specific numbers, such groups might not exist, or at least not in the way the author constructed them. The paper doesn't prove they don't exist, but it strongly suggests an obstruction.

The Bottom Line

This paper solves a puzzle that many mathematicians thought was unsolvable: finding a large, complex group of tree-dancers that stops moving with a non-zero probability, and actually calculating that probability. It provides a clear, explicit formula for these values, turning a vague concept into a precise tool. It also connects this abstract group theory directly to the study of polynomial equations, showing that the symmetries of $x^d + 1$ have a measurable "stopping power."

The author concludes by asking new questions: Can we find groups with any fixed-point proportion we want? What happens if we pick dancers at random? But for now, the main achievement is clear: we now have a factory for these special groups, and we know exactly how many of them stop dancing.