- Rating
- 3
- Confidence
- 3
- Informativeness
- 3
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
Private synthetic data under the Wasserstein distance suffers from the curse of dimensionality. The authors bypass this by considering private sliced Wasserstein guarantees for the data. The technique to achieve these results provide connections between Chebyshev moments and sliced Wasserstein error. The authors also test their method on a variety of experiments.
Critical strengths
- The problem is well-motivated and novel, the paper is generally quite well written, and the proofs are easy to follow.
The strengths generally outweight the weaknesses of the paper.
Critical weaknesses
- In some of the experiments, it seems prior approaches are better at smaller values of , e.g. . The authors should mention this as a limitation. Also the figures do not seem to scale the X axis in a reasonable manner.
- The authors do not provide any lower bounds, which would be useful to see if there are any gaps in their algorithms.
These concerns are minor and addressing them would strengthen the paper in my opinion.
Questions for authors
- Can the authors fix the figures so epsilon is scaled properly (e.g. either exact scale or log-scale), and try smaller values of \epsilon?
- It would be nice if the authors can also provide a lower bound.
2 comments
Thanks for clarifying my concerns. I maintain my score.
Thank you so much for taking the time to review our paper, and for your positive assessment of our work!
Changes made to address your concerns (and the concerns of other reviewers) appear in the updated PDF in blue. We'd also like to highlight that we were able to tighten the main utility theorem after submission, so Theorem 5 now has a dimension-independent rate SW1(p,q)≤O~(1/nϵ) (it was previously O~(d1/4/nϵ)). The improvement is via a relatively minor modification of the original proof given in Appendix C.
This is a fair point. We added a section in the experiments where we highlight a couple of those instances where prior methods are better (for instance, AIM leads on 1-way marginal fidelity at every budget, etc.) However, we note that it is generally true that the methods that lead in these cases are explicitly measuring the statistic being evaluated (e.g., AIM measures low-order marginals directly), while SPLICE spends its budget on sliced moments that target global structure.
Thank you for pointing this out, we appreciate that it was confusing! In the revised PDF, each line plot figure with an ϵ-axis now has a log-scaled x axis where we put the x-axis ticks at the actual privacy budgets.
During the rebuttal period, we ran the full comparison at ϵ∈{0.1,0.25} with the identical protocol. We note that such small values of ϵ are less commonly used in practical DP synthetic data implementations (see e.g. Ponomareva et al. 2026 https://arxiv.org/pdf/2512.03238, section 7.1, where they note that the majority of practical DP synthetic data methods operate in the 1≤ϵ≤10 regime, while ϵ≪1 leads to too poor performance to be practical).
Nevertheless, it is an interesting regime to consider. Below we show SPLICE against the strongest and weakest baselines in this regime (AIM 2-way and Private-GSD), extended to moderate budgets so the trend is visible (means over all datasets covered by every shown method at every budget, 43 for TVD and 36 for SW1; lower is better):
At the tightest budgets SPLICE trails the marginal-based methods: as SPLICE spreads its budget over Lk sliced moments, at very small ϵ the per-moment noise swamps the available signal (though even at ϵ=0.1, every method still beats the trivial Dirac baseline discussed with Reviewer h6QY, whose mean SW1 error is 0.198 on this set). The table also shows how quickly this reverses: SPLICE improves roughly 7x on SW1 from ϵ=0.1 to 3 and overtakes every baseline on both metrics by ϵ=1. In our revision, we have added the small-ϵ results for all seven methods to the appendix (Table 3, "The tight-budget regime" in Appendix G), and refer to it in the new Limitations paragraph in the main body.
This is a good question. We have added a remark on lower-bounds in the revision (see the remark in Appendix C, beside our trivial upper bound lemma). In summary, we note the following. For d=1, the SW1=W1, and a known lower bound of Ω~(1/(ϵn)) holds for DP synthetic data matching any distribution; this floor is tight, as there are a few known methods that match it (e.g. Musco et al. 2025). For general d, this implies a lower bound of Ω(1/(ϵnd)), which follows from embedding a difficult one-dimensional distributional instance along a line that passes through the origin. The sliced Wasserstein distance shrinks by Θ(1/d) (the expected inner product of a random direction and the direction of that line). So, for constant d, we cannot hope to obtain a bound better than O(1/(nϵ)), which is better than our current bound of O~(1/nϵ). Whether or not the square root can be removed, or the lower bound improved in d>1 dimensions, is a nice open theoretical question.
We thank you again for your time and attention to our work, we believe we have addressed your questions/concerns but welcome a continued discussion.