WorksMathematics
Iterated Steiner symmetrization of convex bodies always converges in shape
Abstract
We prove that every sequence of iterated Steiner symmetrals of a convex body converges in shape. That is, every such sequence can be corrected by a sequence of rotations to achieve convergence in Hausdorff distance. Moreover, for a fixed corresponding family of subspaces, the same rotations can be used for every initial convex body, and their successive increments can be chosen to be square-summable.
Introduction
The story of Steiner symmetrization started almost two centuries ago with the work of Jakob Steiner on the isoperimetric inequality for convex bodies (that is, non-empty compact and convex subsets of ), based on the following remarkable property: Steiner symmetrization preserves the Lebesgue measure of a convex body while decreasing its surface area. Moreover, given a convex body , it is always possible to find a sequence of hyperplanes such that the corresponding iterated Steiner symmetrals converge to a centred Euclidean ball in the Hausdorff metric. For more details, we refer the reader to the exposition by Bianchi and Gronchi [5].
The first general result in is attributed to Gross [9], who proved that for every convex body there exists a symmetrization process converging to a centred Euclidean ball. Decades later, Mani-Levitska [11] proved a much stronger statement: Choosing the sequence of hyperplanes independently and uniformly at random, and applying the corresponding symmetrization process to a convex body with non-empty interior, the corresponding sequence of symmetrals will converge almost surely to a centred Euclidean ball. See Burchard and Fortier [7] and Coupier and Davydov [8] for further results on convergence of random Steiner symmetrizations. Even if these results do not leave too much space to non-converging symmetrization processes, a counterexample was eventually found in the work of Bianchi, Burchard, Gronchi, and Volčič [1]. In the same work, the notion of convergence in shape was introduced in order to fix the counterexample: A sequence of convex bodies is said to converge in shape if there exists a sequence of rotations such that the adjusted sequence converges in the Hausdorff metric. In [1 Theorem 2.2], it is then proved that a class of Steiner symmetrization processes which includes the aforementioned counterexample does indeed always converge in shape. In that case, the successive angles between the symmetrizing directions are square-summable, and the same rotations work for all initial compact sets.
It is natural to ask whether every sequence of Steiner symmetrals converges in shape, a problem posed in Question 6.2 of [1]. In this paper, we give a positive answer in the class of convex bodies. Additionally, inspired by a conjecture of Bianchi and Gronchi [4], we show that
MSC 2020 Classification: Primary 52A20. Secondary 51F15, 52A30. every non-convergent sequence nevertheless admits correcting rotations with square-summable successive increments, a property shared by the counterexample [1]. In fact, for a given sequence of hyperplanes, these rotations can be chosen once for all initial convex bodies.
This work is completed by some consequences and extensions to more general symmetrizations, a framework introduced by Bianchi, Gardner, and Gronchi [2, 3]. In particular, our results extend to Minkowski symmetrization which, in turn, allows us to analogously treat compact subsets of , following a procedure from [14].
Main result. We start by introducing some notation. We denote by the family of convex bodies in and by the linear hyperplanes in . For and the corresponding Steiner symmetral (see § 2 for the precise definition of Steiner symmetrization). For a sequence , the corresponding Steiner symmetrization process produces, for every , the iterated sequence of symmetrals
Convergence of convex bodies is always understood with respect to the Hausdorff distance, denoted by . Moreover, we denote by the usual distance on induced by its standard Riemannian structure (see § 3 for the details). Our main result reads as follows.
Theorem 1.1. Consider a sequence of hyperplanes . Then there exists a sequence of rotations such that, for every , the sequence of symmetrals , , satisfies
for some . The rotations can be chosen with square-summable successive increments. That is,
Additionally, if does not converge for some , then these rotations satisfy
Notice that if converges for every , one can trivially choose , the identity matrix, for every . Therefore, Theorem 1.1 can be understood as the following dichotomy: Either the symmetrization process converges for every initial body, or all its sequences of symmetrals converge in shape with the same rotations satisfying (1). While the proof of convergence in shape for a single body uses basic theory of convex bodies, the summability statement and the choice of a common sequence of rotations draw on additional elementary tools from the theory of Lie groups and spherical harmonics. For this reason, we correspondingly split the proof of Theorem 1.1 between § 2 and § 3.
Structure of the paper. In § 2, we introduce further notation and definitions, upon which we build the proof of convergence in shape for a single body. Moreover, we show how (1) adapts to the case of ellipses in the plane, providing a preliminary intuition of the convergence mechanism. § 3 is dedicated to proving that condition in complete generality and to choosing the rotations independently of the initial body. Finally, in § 4, we show how the core constructions apply to the more general framework of symmetrizations introduced in [2, 3].
Convergence in shape
We recall that by we denote the family of non-empty compact and convex subsets of , known as convex bodies. More generally, denotes the non-empty compact subsets of . For , its support function is , , where is the Euclidean unit sphere of and is the standard scalar product in . Both and are endowed with the topology induced by the Hausdorff metric. See Schneider’s monograph [12] for further details on the theory of convex bodies.
Consider and . The Steiner symmetral is obtained by replacing every non-empty interval , , by the (possibly degenerate) interval of the same length centred at and orthogonal to . The resulting set is again compact and convex. We only use Steiner symmetrization on below. The final compact-set extension concerns Minkowski symmetrization.
For , their Minkowski sum is the set . Moreover, for we denote by the linear reflection in with respect to . A subset of is -symmetric if it is invariant for . For and , the Minkowski symmetral of with respect to is the compact set . If, additionally, is convex, then so is . When restricted to , Steiner and Minkowski symmetrizations satisfy the inclusion
See, for example, [8 Lemma 2.1]. In Section 4, we will explain how the generalization of this inclusion in the framework of [2, 3] allows us to extend our results.
The aim of this section is to prove convergence in shape for a fixed initial body, which we state here for the convenience of the reader.
Theorem 2.1. Consider , a sequence , and the corresponding sequence of Steiner symmetrals
Then, there exists a sequence of rotations and (both allowed to depend on and ) such that
The result above has the following immediate consequence.
Corollary 2.2. All subsequential limits of are the same up to a rotation in .
Proof of Theorem 2.1
We start with some preliminary observations. First, we recall that given and , choosing large enough so that , where is the Euclidean unit ball, the sets satisfy for every since Steiner symmetrization preserves inclusions. For the remainder of this subsection the set , the sequence , and the corresponding symmetrization process will be fixed.
In terms of support functions, writing for , we infer
for every O(n)$ the orthogonal linear group of , for and we write . Consider now the transformations
They are self-adjoint orthogonal projections on and hence 1-Lipschitz. Moreover, (2) implies
Indeed, by the definition of Minkowski symmetrization,
Thus (3) follows from (2). Denote by the standard scalar product on and by the corresponding norm. From (3) and since is 1-Lipschitz, we infer
In particular, the sequence converges to a value since it is non-increasing.
Next, we introduce the following pseudo-distance, which will be the key ingredient of the proof of Theorem 2.1. For , consider
It is well defined since is easily checked to be continuous with respect to the standard topology of (see § 3 for more details) and is compact. In particular, . For the sequence of symmetrals , satisfies the following property.
Lemma 2.3. Given the sequences and as above, for every $j>i\geq 1
Proof. Fix . Repeated use of (3) and preservation of pointwise order gives
Expanding the operators, we can write
where the are compositions of reflections. Hence
Since this is an average of nonnegative numbers, at least one term satisfies the same upper bound. Thus, for some ,
If , replace by . This does not change because is -symmetric. The last display is therefore an admissible competitor for and proves (4). □
:::
Next, observe that for every
Indeed, notice that for
In particular,
and (5) is proved inverting the role of and .
To finish the proof of Theorem 2.1, by the Blaschke selection theorem [12 Theorem 1.8.7] there exists a subsequence converging to some in the Hausdorff metric. For every apply Lemma 2.3 to the indices and let to obtain
The right-hand side tends to zero. Now, for every , choose a rotation minimizing . Then in . Since all these convex bodies lie in the fixed ball , convergence in the support-function metric is equivalent to Hausdorff convergence, see [12 Notes for Section 1.8, Note 9]. Hence .
A 2-dimensional example: Behaviour on ellipses
Before completing the proof of Theorem 1.1 in the next section, we present an easier proof restricted to centred ellipses in to give a geometric intuition of (1), which we are going to prove in full generality in the next section. We recall that for every 1-dimensional subspace , the set is still an ellipse, and its axes are aligned along and .
Theorem 2.4. Consider a sequence . Then, for every centred ellipse with semiaxes of lengths , the sequence of symmetrals
has necessarily one of the following mutually exclusive behaviours:
The sequence converges to some ellipse .
The sequence does not converge, but converges in shape to some ellipse and, denoting by the smaller angle between the major axes of and ,
Proof. If converges, then we are in the first case. Suppose otherwise. Theorem 2.1 implies convergence in shape. Write and , where are the semiaxis lengths. We represent a centred ellipse by a positive definite matrix through . In coordinates in which is the first axis, an ellipse represented by is sent by to the ellipse represented by . Consequently, is nonincreasing, is nondecreasing, and . Hence and , with . The inequality is strict, since otherwise would force convergence to a disk. No symmetral $E_m can itself be a disk, since all subsequent symmetrals would then be the same. Thus, all the major axes are well defined. Finally,
because a finite sum would make the sequence of major-axis lines a Cauchy sequence in the angular metric. Together with convergence of the semiaxes, this would imply convergence of .
Now, observe that
Indeed, for two successive symmetrals and , the major axis shifts in one of the two following ways: Either it aligns with , in which case
or with , and
In both cases, we obtain (6). To conclude, the elementary inequality for , together with (6), gives
Thus,
Choosing the right rotations in Theorem 1.1
To obtain the full statement of Theorem 1.1, we first work with a single sequence of symmetrals. The main idea is the following: Working on the orbit of the possible limits of the sequence , one can iteratively choose the rotation from Theorem 2.1 in a suitably small neighbourhood of inside . To this end, we work on as a Lie group and a Riemannian manifold. We start by recalling the corresponding metric. In this section, for ease of notation we write . Then, on we consider its standard Riemannian metric, which, at the identity, reads as
Here, is the Lie algebra of . See, for example, [6] for this and other basic definitions. As it is standard, the metric on the whole is defined by left translates. Recall that both left and right translates are isometries for this metric. In particular, for every
We will need the following lemma. Given a real finite-dimensional inner-product space , denotes its orthogonal group and its norm.
Lemma 3.1. Consider an orthogonal smooth representation of on . That is, we are given a smooth group homomorphism
For a fixed consider its orbit
Then is a smooth embedded submanifold of . Moreover, there exist , , such that for every with there exists satisfying
Proof. First, suppose that . Then the statement is trivial. Therefore, in the remainder of the proof we suppose that is not a singleton. Since is compact, its smooth action on is proper. See, e.g., [10 Corollary 21.6]. As orbits of proper Lie group actions are embedded submanifolds, so is , which is compact since it is the continuous image of . This proves the first claim.
To prove the second part observe that the orbit map
is a smooth submersion. By the local section theorem [10 Theorem 4.26] there is a smooth local right inverse of with . Then, upon choosing sufficiently small, there exists a constant such that for every with holds
The conclusion now follows by a standard translation argument. For with , choose with and set . Then satisfies
where bi-invariance of and orthogonality of the representation were used.
Square-summable convergence in shape
We now prove that the rotations in Theorem 2.1 can be chosen to have successive square-summable distance in . This behaviour already appeared in [1], and Bianchi and Gronchi [5] suggested the author that this phenomenon should extend to all sequences of iterated Steiner symmetrals of convex bodies.
Theorem 3.2. Consider as in Theorem 1.1. If does not converge, then for any subsequential limit of the rotations giving can be chosen so that
Proof. Retain the notation , , , from § 2, and fix a limit as in the proof of Theorem 2.1. We start by refining the main argument in the proof of Lemma 2.3. Since is self-adjoint and ,
Consequently,
and therefore
Thus, . Write and consider the compact orbit
By the considerations above, . Although is homeomorphic to the finite-dimensional homogeneous space , its linear span may be infinite-dimensional. In order to be able to apply Lemma 3.1, we now get to a reduction argument.
For let be the space of spherical harmonics of degree and set
These orthogonal projections commute with rotations and satisfy in . Define
Each is a closed subgroup of . Since the projections commute with rotations and , these groups form a decreasing sequence. The descending-chain theorem for closed subgroups of a compact Lie group [6 p. 136] gives such that for every . Moreover,
because for in the intersection, . Therefore
Write , , and . Equality of stabilizers implies that is injective on the orbit . Let
Then and .
If is a singleton, injectivity of makes a singleton, and would imply convergence of , contrary to the hypothesis. Thus is not a singleton. By Lemma 3.1, is a compact embedded submanifold of . By the tubular neighbourhood theorem [10 Theorem 6.24], nearest-point projection onto is smooth on a sufficiently small tube, and its differential is bounded on a smaller closed tube by compactness. Since and , the segment from to lies in the smaller tube for all sufficiently large . In particular, the points satisfy
Choose sufficiently large that Lemma 3.1 gives for such that
Additionally, consider with . Set for , and for define
Then , and
Finally, we show that . Put . Since commutes with rotations,
Also,
Since is compact, every subsequence of has a further subsequence converging in to some , and (11) implies . Since is injective, . Hence in .
Since
the equivalence of the support-function and Hausdorff topologies on uniformly bounded convex bodies [12 Notes for Section 1.8, Note 9] yields .
Remark 3.3. We use a dimension reduction from to , since even if is finite dimensional, its linear span need not be. E.g. support functions of noncircular ellipses have infinitely many Fourier coefficients, so their rotation orbits have infinite-dimensional linear span.
Finding a common sequence of rotations
We are left to show that the rotations in Theorem 1.1 can be chosen independently of the initial body. We start by considering finitely many initial bodies and applying the preceding arguments to their symmetrals simultaneously. For a tuple and , we write . Convergence of tuples is understood in the Hausdorff metric in each component.
Lemma 3.4. Consider a sequence and a finite family , with . For each let , and write . Then any two subsequential limits of differ by a common rotation. Moreover, for each subsequential limit there is a sequence satisfying (8) such that .
Proof. Choose such that for every . In , endowed with the product scalar product, put
We let and act in each component as in § 2. By (4), componentwise, so decreases to a limit . Notice that in the expansion of the same compositions of reflections appear in each component. Moreover, if , then and , since each is -symmetric. Thus the proof of Lemma 2.3 gives, for ,
Now let be a subsequential limit and put . Passing to the corresponding subsequence in the preceding inequality gives
In particular, since is compact, every subsequential limit of lies in . The first assertion follows since support functions determine their bodies.
If converges, we can take . Otherwise, notice that . As in the proof of Theorem 3.2, we then obtain
Apply in every component. The stabilizers of the resulting projections of form a decreasing sequence with intersection . By the descending-chain theorem, the projection onto is therefore injective on for some . We can now repeat the nearest-point projection and lifting argument of Theorem 3.2 in . It gives rotations satisfying (8) and in . By the equivalence of the support-function and Hausdorff topologies on uniformly bounded convex bodies, .
We now choose a finite family for which adjoining another sequence of symmetrals cannot reduce the stabilizer of the limit tuple. For its stabilizer is .
Proposition 3.5. For every sequence there exists a sequence satisfying (8) such that, for every , the symmetrals satisfy in the Hausdorff metric for some .
Proof. Fix and consider the collection of the stabilizers of all subsequential limits of tuples as in Lemma 3.4. Since these are closed subgroups of , the descending-chain theorem gives a minimal member under inclusion. Choose a tuple of symmetrals and a subsequential limit with . By Lemma 3.4, we can fix rotations satisfying (8) such that .
Consider now any other initial body and let be its sequence of symmetrals with respect to the fixed family . By the Blaschke selection theorem, along a subsequence on which we may assume that for some . We observe that
By minimality of , this intersection equals . In particular, every fixes .
Let be a limit of a subsequence of . By compactness we may pass to a further subsequence on which
for some tuple of convex bodies and a further . Applying Lemma 3.4 to the enlarged tuple, we find such that . Since , we have , and consequently . The corrected symmetrals are uniformly bounded, so uniqueness of their subsequential limit gives . This proves the result since the rotations were fixed before choosing .
Completing the proof of Theorem 1.1. By Proposition 3.5, there is a common sequence satisfying (8). If the symmetrals converge for every initial body, we may take . Suppose instead that there is an initial body for which fails to converge. If , the rotations are Cauchy and converge to some . But then implies , a contradiction. This completes the proof.
Final remarks and extensions
Apart from -symmetry of the symmetral, a careful inspection of the arguments shows that the key properties of Steiner symmetrization which are used to prove the results above are the following:
a) For every such that , ,
b) If satisfy then ,
c) For every the inclusion holds.
In fact, (a) and (b) were used to keep the symmetrals in a fixed ball, and this also follows directly from (c) since is monotone and fixes . It follows that if we were to identify maps of into itself producing -symmetric sets and satisfying (c), the proofs in the previous sections would apply to these maps as well. For example, Minkowski symmetrization satisfies the required assumptions. As a matter of fact, Bianchi, Gardner, and Gronchi [2 Corollary 7.3] proved that (a) and (b) give (c) under one additional condition. These maps are particular examples of -symmetrizations, as defined in [2, 3]. We report a version of [2 Corollary 7.3] adapted to our context. Another application of this result to convergence of symmetrization processes can be found in [13].
For a linear subspace of , we use the same notation for the orthogonal map that fixes and acts by on . The Minkowski symmetral is defined by the same formula as in Section 2. We denote by the fibre symmetral of , obtained by replacing each non-empty section , , by its Minkowski symmetral within that fibre with respect to reflection about . In particular, when is a hyperplane.
Proposition 4.1. Consider a map
such that for every and one has . Suppose, moreover, that the analogues of (a) and (b) hold with replaced by , and that
whenever is -symmetric and . Then
for every such and every .
The corresponding convergence result is as follows. As in Theorem 1.1, we allow the rotations to depend on the sequence of subspaces alone.
Corollary 4.2. Consider a sequence of proper linear subspaces of , possibly of different dimensions. Then there exists a sequence satisfying (9) with the following property. For every and every sequence of -symmetrizations such that for every , the sequence
satisfies
for some . In particular, the same rotations apply to fibre and Minkowski symmetrizations and to the symmetrizations in Proposition 4.1. If one of these uncorrected sequences does not converge, the rotations satisfy (1).
Proof. The containment in the Minkowski symmetral implies (4), and -symmetry gives . These observations still hold for a tuple of processes that use different symmetrizations along the same subspaces. When each is a hyperplane, the proof of Theorem 1.1 therefore applies without further changes.
For arbitrary subspaces we need to reconsider the last step in the proof of Lemma 2.3, since can have determinant . Before this step, the proof gives (5) with the minimum over , also for tuples. As in the proof of Theorem 2.1, it follows that approaches the -orbit of the support-function tuple of any subsequential limit . This orbit consists of at most two -orbits. If there are two, they are compact and disjoint, hence have positive distance. On the other hand, (10) holds in the product norm and shows that . Thus, for sufficiently large , the tuple cannot pass from a small neighbourhood of one orbit to a small neighbourhood of the other. Since is a subsequential limit, approaches the -orbit corresponding to .
We can now use the proof of Theorem 3.2 for tuples, followed by the minimal-stabilizer argument in § 3.2. In choosing a minimal stabilizer there, we consider the tuple limits of all the processes in the statement, so that the resulting rotations are common to them all. The same argument at the end of § 3 gives the first half of (1) if any uncorrected sequence fails to converge.
We conclude with a further extension that applies to Minkowski symmetrization. As pointed out in [14], the behaviour of Minkowski symmetrization on can be used to describe the one on the family of non-empty compact subsets of , which we denote by . The corresponding result in this case looks as follows.
Corollary 4.3. For and a sequence as in Corollary 4.2, consider the sequence
Then the same rotations as in Corollary 4.2 give
for some . In particular, one sequence of rotations works for every . If does not converge for some , these rotations satisfy (1).
Proof. Put . First, recall that the convex hull commutes with Minkowski addition and, therefore, with Minkowski symmetrization. In particular, is the Minkowski symmetrization process starting from . Moreover, the estimate from the Shapley–Folkman–Starr theorem [12 Theorem 3.1.6] gives
See [14 Theorem 3.4 and Corollary 3.5] for a similar argument. Applying the Corollary 4.2 to and using rotation invariance of , we obtain convergence of to the same limit as . Finally, the estimate shows that converges if and only if converges, so the last claim also follows Corollary 4.2. □
Acknowledgments. The author was supported, in part, by the Austrian Science Fund (FWF): 10.55776/PAT3787224. The author thanks Georg Hofstätter for insightful discussions.
AI disclosure. This manuscript, both in writing and mathematical proofs, was prepared with the assistance of AI. In particular, GPT-6 Astra was used. The mathematical results were assisted and refined through various rounds of prompting. The final form of proofs and writing in the manuscript were handcrafted by the author, and double-checked with AI for typos and stylistic suggestions.
Jacopo Ulivelli
Institut für Diskrete Mathematik und Geometrie
TU Wien
Wiedner Hauptstraße 8–10/1046
1040 Wien, Austria
e-mail: [email protected]
References
- [1]G. Bianchi, A. Burchard, P. Gronchi, and A. Volčič, Convergence in shape of Steiner symmetrizations, Indiana Univ. Math. J. 61 (2012), pp. 1695–1710.DOI
- [2]G. Bianchi, R. J. Gardner, and P. Gronchi, Symmetrization in geometry, Adv. Math. 306 (2017), pp. 51–88.DOI
- [3]G. Bianchi, R. J. Gardner, and P. Gronchi, Convergence of symmetrization processes, Indiana Univ. Math. J. 71 (2022), pp. 785–817.DOI
- [4]G. Bianchi and P. Gronchi, private communication, 2019.
- [5]G. Bianchi and P. Gronchi, Symmetrizations, in Convex Geometry, Lecture Notes in Math., vol. 2332, Springer, Cham, 2023, pp. 233–292.DOI
- [6]T. Bröcker and T. tom Dieck, Representations of Compact Lie Groups, Graduate Texts in Mathematics, vol. 98, Springer-Verlag, New York, 1995, corrected reprint of the 1985 translation.
- [7]A. Burchard and M. Fortier, Random polarizations, Adv. Math. 234 (2013), pp. 550–573.arxiv.org/abs/1104.4103
- [8]D. Coupier and Y. Davydov, Random symmetrizations of convex bodies, Adv. in Appl. Probab. 46 (2014), pp. 603–621.arxiv.org/abs/1211.1785
- [9]W. Groß, Die Minimaleigenschaft der Kugel, Monatsh. Math. Phys. 28 (1917), pp. 77–97.
- [10]J. M. Lee, Introduction to Smooth Manifolds, Graduate Texts in Mathematics, vol. 218, 2nd ed., Springer, New York, 2013.
- [11]P. Mani-Levitska, Random Steiner symmetrizations, Studia Sci. Math. Hungar. 21 (1986), pp. 373–378.
- [12]R. Schneider, Cover Bodies: The Brunn–Minkowski Theory, Encyclopedia of Mathematics and its Applications, vol. 151, 2nd expanded ed., Cambridge University Press, Cambridge, 2014.
- [13]J. Ulivelli, Convergence properties of symmetrization processes, Adv. in Appl. Math. 146 (2023), Paper No. 102484, 18 pp.DOI
- [14]J. Ulivelli, Generalization of Klain’s theorem to Minkowski symmetrization of compact sets and related topics, Canad. Math. Bull. 66 (2023), pp. 124–141.