- Rating
- 2
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
This paper studies the complexity of the classical Moreau-Yosida unadjusted Langevin algorithm (MYULA) for sampling from strongly log-concave distributions with a smooth strongly convex component and a nonsmooth convex Lipschitz component. Building on the Poisson-corrector framework and the active-trace analysis, the authors establish a refined invariant-measure bias bound of order , with only logarithmic dependence on the inverse smoothing parameter in the error coefficients. Combining this estimate with the Moreau approximation bias yields an end-to-end iteration complexity of , improving upon the bound obtained by directly applying the existing Poisson-corrector result.
Critical strengths
- The paper addresses a relevant theoretical problem concerning the interaction between Moreau regularization and discretization errors in Langevin sampling for nonsmooth potentials.
- The improved complexity bound is a meaningful quantitative improvement over the benchmark. The result applies to general convex Lipschitz nonsmooth components without requiring additional geometric assumptions on their structure.
- The shared-noise two-point curvature estimate provides an interesting technical ingredient for controlling the dependence on the smoothing parameter.
Critical weaknesses
-
The analysis largely follows the Poisson-corrector framework of Pedrotti and Whalley (2026) and builds on the active-trace techniques of Xin and Zhang (2026). The main new ingredient is the shared-noise two-point curvature estimate, together with refined moment and variance bounds. While these estimates appear technically nontrivial, the contribution is primarily a refinement of existing analytical techniques. It remains unclear whether the proposed analysis provides broader theoretical insights beyond improving the complexity bound for classical MYULA.
-
Although the improved rate is meaningful, the improvement is established asymptotically with fixed model parameters and dimension. The explicit bounds also introduce a potentially less favorable dimension dependence than the direct Poisson-corrector benchmark. A more detailed comparison of the parameter regimes in which the new result provides a substantial improvement would strengthen its significance. Furthermore, it remains unclear whether the bias dependence reflects an intrinsic feature of MYULA or a limitation of the current analysis.
Questions for authors
-
The proposed analysis builds on the Poisson-corrector framework of Pedrotti and Whalley (2026) and the active-trace approach of Xin and Zhang (2026). Could the authors clarify what fundamental mathematical difficulty is resolved by the new shared-noise two-point curvature estimate, and whether this estimate provides insights beyond the existing frameworks?
-
The proposed invariant-measure bias bound scales as , whereas the existing Poisson-corrector analysis achieves for a fixed smooth potential, albeit with unfavorable dependence on the smoothing parameter. Is the dependence inherent to obtaining favorable smoothing dependence, or does it arise from specific inequalities in the current analysis? What are the main obstacles to improving this rate?
-
Can the shared-noise curvature estimate be extended to other Langevin discretizations or nonsmooth sampling algorithms? If not, which properties of the classical MYULA dynamics are essential to the argument?
3 comments
Part 2/2
3.Origin and possible improvement of the rate
Where the exponent comes from.
The rate arises from our current estimates; we do not establish it as an inherent cost of favorable smoothing dependence. The nonsmooth martingale variance and curvature feedback each contribute per step (Propositions 6.2–6.3). Accumulation under contraction gives mean-square error, hence the bound (Section 6.4).
Possible limitation of synchronous coupling.
Figures 1–2 support synchronous RMS scaling along the tested decreasing-smoothing paths, while suggesting Wasserstein bias of order . As discussed at the end of Section 7, this suggests that our analysis captures the synchronous-error scaling, but bounding Wasserstein bias by this error may limit the rate.
Obstacles to improvement.
Within the current recursion, improving only one nonsmooth estimate leaves the other as a bottleneck. Moreover, if the observed synchronous-error scaling persists asymptotically, sharpening the RMS bound for this same coupling would not suffice to obtain an Wasserstein bound. Such an improvement may require exploiting cancellations at the level of the invariant laws or using a different coupling, while avoiding the unfavorable inverse-power dependence on in the existing fixed-smooth-potential analysis.
4.Parameter regimes and dimension dependence
We have added a parameter comparison in Section 2 (“Poisson correctors and a direct MYULA benchmark”), complementing our main improvement of the accuracy exponent from to for fixed model parameters and dimension.
Dimension dependence and the role of .
For , our bound has a larger dimension factor than the benchmark, but improves it when , up to logarithmic factors. Dimension dependence can also improve when grows with . For example, with fixed has . With , our bound becomes , compared with for the benchmark, improving both exponents. Thus, the comparison depends jointly on accuracy, dimension, and model parameters such as ; neither bound is uniformly better across all parameter regimes.
Thank you for the clarifications. However, my main concern remains. The main contribution seems to be sharper estimates within the existing Poisson-corrector framework. The dimension dependence is worse in some regimes, and the numerical results suggest that the Wasserstein bound may not be sharp. Although the improved complexity rate is useful, I still find the overall contribution rather limited and therefore maintain my original score.
Part 1/2
Thank you for your comments and questions. We have revised the manuscript, while the main theoretical results and proof arguments remain unchanged in substance. All references below refer to the first revised manuscript, iclr2027_R1.
1. Mathematical difficulty and the new estimate.
The corrected-error recursion in Section 6.1 requires controlling E[χh(Xkh)⊤Akgχh(Xkh)], where χh is the Poisson corrector and Akg represents the Moreau curvature between the coupled states. Using the worst-case bound Akg⪯λ−1I gives h2R2/λ, whereas Proposition 6.3 gives O(h3/2), with only logarithmic dependence on λ−1. As explained in Section 6.3, this changes the contribution of this particular term to the Wasserstein bound from O(h/λ) to O(h3/4).
The difficulty is that the relevant curvature lies along the random segment joining the actual diffusion and Euler states, which is not directly controlled by existing reference-path averages. Proposition D.1 uses their shared Gaussian noise to average curvature along this segment and control it through the bounded Moreau gradient, avoiding the worst-case curvature bound. Combined with higher corrector moments and Hölder’s inequality (Appendix D.2), this handles the correlation between curvature and the corrector and improves the smoothing dependence of this term within the existing Poisson-corrector framework.
2. Generality and broader theoretical insights
Generality of the curvature estimate.
The mechanism in Section 6.3 and Appendix D.1 is that a common Gaussian shift of two endpoints also shifts the entire segment joining them. For a convex function with bounded gradient, Gaussian integration by parts then controls the expected Hessian trace along this segment through the gradient bound(see Appendix D.1, proof of Proposition D.1, Step 1). For the exact diffusion, the endpoint is not a simple Gaussian shift because the drift also depends on the noise. Our small-step condition ensures that its response to the shared noise remains close to that of the Euler endpoint, allowing the same averaging argument to apply. These conditions are not specific to MYULA: for example, replacing the deterministic Euler update by another deterministic map before adding the same Gaussian noise leaves this local argument unchanged.
Broader theoretical insight and scope.
The broader insight is that shared noise can control curvature along the segment between the actual coupled states, even when the worst-case Hessian bound diverges as smoothing decreases. This may provide a reusable way to exploit averaged curvature in coupling analyses. Whether it improves another algorithm’s convergence rate requires further analysis: the corresponding curvature term must arise and limit the existing bound, while the remaining local-error, corrector, and mixing estimates must also be established.