REVIEW
On unboundedness of maximal operators for directional Hilbert transforms
Classical Analysis and ODEs · Submitted 23 Feb 2006 · 20 references
Pith's verdict: Settles the natural open question—every infinite direction set gives an unbounded maximal directional Hilbert transform—but the printed proof of Theorem 2 has two repairable gaps that a referee should catch.
T0 · AI review
Why Pith says this
Highlight any text to comment on it.
Referee report
The paper
The paper claims
For any infinite set of directions in the plane, the maximal operator that takes the supremum of one-dimensional Hilbert transforms is unbounded on $L^2(\mathbb{R}^2)$.
The reading
What carries the argument
The load-bearing object is a finite tree-system of functions: each function is supported on the set where a previous function in a binary nesting has a fixed sign. Lemma 3, adapted from a classical lemma for the Haar system, says that any such tree-system can be permuted so that, pointwise, the largest initial partial sum of the permuted functions is at least one third of the sum of their absolute values. The proof of Theorem 2 builds a tree-system from highly oscillatory functions with Fourier support in the sectors $S_k$ between the perpendiculars to consecutive ordered directions. The directional Hilbert transform turns each sector piece into a single term $T_{S_k}f$, so the maximal directional transform becomes the supremum of partial sums of these sector terms, and the tree-system lemma supplies a large value on a positive-measure set. The identity that identifies the supremum over directions with the maximum of initial sector sums is the step that carries the whole argument.
Extended reading notes
Core claim
The paper's central claim is that for any infinite set of unit vectors $U\subset\mathbb{R}^2$, the operator $H_U f(x)=\sup_{u\in U}|\mathrm{p.v.}\int f(x-tu)\,dt/t|$ cannot be extended to a bounded operator from $L^2(\mathbb{R}^2)$ to $L^p(\mathbb{R}^2)$ for any $1\le p<\infty$. This is derived from a finite-set estimate: for every finite set of $N$ directions, $\|H_U\|_{L^2\to L^p}\ge c\sqrt{\log N}$ with an absolute constant $c>0$. Because this lower bound tends to infinity with $N$, no infinite set of directions can admit a uniform bound. The proof constructs a test function whose Fourier transform is supported in narrow sectors between consecutive directions, arranges the sector pieces as a tree-system, and uses a permutation lemma to force one initial partial sum of directional Hilbert transforms to exceed $c\sqrt{\log N}$ on a set of positive measure.
Load-bearing premise
The proof's lower bound stands on the assumption that, after ordering the directions inside one quadrant, the maximal half-plane projection is exactly the supremum of the initial partial sums of the sector projections; without this identity the constructed function may not satisfy the claimed bound.
Editorial extensions
If this is right
- For any infinite set of directions $U$, $H_U$ fails to be bounded from $L^2(\mathbb{R}^2)$ to $L^p(\mathbb{R}^2)$ for every finite $p\ge 1$.
- Any finite set of $N$ directions satisfies a lower bound of order $\sqrt{\log N}$, so the operator norm must grow at least at that rate as directions are added.
- The previously open case of infinite lacunary families of directions is resolved: even these sparse families admit no bounded maximal Hilbert transform on $L^2$.
- Since each single-direction Hilbert transform is bounded on $L^p$, the unboundedness is a phenomenon of the supremum over directions rather than of the individual transforms.
Reading between the lines
- The identity (34) is not correct as written: for the sectors defined in (33), the half-plane of a given direction corresponds to a tail sum of sector terms, not an initial sum. A reader could repair the estimate by comparing tail sums instead, since the largest tail sum is at least half the largest partial sum, so the theorem is likely salvageable with a small modification.
- The tree-system construction is combinatorial in character, which suggests the same mechanism may give lower bounds for directional maximal operators in higher dimensions, where sectors become cones and the permutation ordering may still apply.
- One could test whether the lower bound $\sqrt{\log N}$ can be improved toward the known $\log N$ upper bound, or whether the upper bound can be sharpened; the paper leaves the gap between the two rates open.
Original abstract
We show that for any infinite set of unit vectors $U$ in $\ZR^2$ the maximal operator defined by $$ H_Uf(x)=\sup_{u\in U}\bigg|\pv\int_{-\infty}^\infty \frac{f(x-tu)}{t}dt\bigg|,\quad x\in \ZR^2, $$ is not bounded in $L^2(\ZR^2)$.
The compiled record
The paper's own sentences, arranged, kept beside the review and separate from it. Each label is drawn from a closed vocabulary; the number beside it is the probability the classifier gave that label. Numbers in the tables are the authors', checked by code. No sentence in this record was written by a model.
The abstract, sentence by sentence
We show that for any infinite set of unit vectors $U$ in $\ZR^2$ the maximal operator defined by $$ H_Uf(x)=\sup_{u\in U}\bigg|\pv\int_{-\infty}^\infty \frac{f(x-tu)}{t}dt\bigg|,\quad x\in \ZR^2, $$ is not bounded in $L^2(\ZR^2)$.
background contribution method result interpretation main claim
| # | sentence | role | the authors' own work | states the main conclusion | strength |
|---|---|---|---|---|---|
| s0 | We show that for any infinite set of unit vectors $U$ in $\ZR^2$ the maximal operator defined by $$ H_Uf(x)=\sup_{u\in U}\bigg|\pv\int_{-\infty}^\infty \frac{f(x-tu)}{t}dt\bigg|,\quad x\in \ZR^2, $$ is not bounded in $L^2(\ZR^2)$. | result 0.96 | 0.96 | 0.96 | asserted 1.00 |
What was examined, and what was not
- arXiv listing abstract v2
- full text
- figures and tables
- references and citing sentences
- relations to other papers
- P abstract
- S sections
- T body sentences
- C citations
- R cross-paper relations
References
Discussion
Disagree with the review, or know something it missed? Leave a comment, highlight any passage above to comment on it, or challenge this review formally.