- Rating
- 3
- Confidence
- 3
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 3
Read review
Paper summary
The paper studies stochastic monotone variational inequalities and provides new convergence analyses for two methods using Halpern-type anchoring under different oracle models. Motivated by learning in games with one stochastic gradient observation per round, the first method uses one oracle call per iteration. In the constrained setting, the analysis improves the anytime last-iterate convergence rate for the restricted gap from to , and establishes the same rate for the gradient mapping norm. The main insight in the analysis is to transfer deterministic convergence bounds to stochastic iterates using a comparison of their trajectories and the Lipschitz continuity of the gradient mapping.
Moreover, the paper studies a second setting, permitting two oracle evaluations at different points using the same sample, the authors analyze a method incorporating STORM variance reduction. Assuming mean-square smoothness, they prove an anytime last-iterate residual rate.
Critical strengths
The authors address an important class of problems and obtain improved convergence rates over those existing in the literature.
Critical weaknesses
There are some minor mistakes in the math / proofs. None of them break the theoretical results, but they do harm readability of the work, and should thus be addressed.
- Proof of Lemma B.3: above (B.11) the authors divide by . Algorithm 2 states that , in which case this would lead to a division by 0. Probably, this is fixed by simply restricting , or handling separately, but this should be addressed nonetheless.
- Bottom of page 5: is assumed, but this does not imply that for all as required for the theory. I suppose is assumed to be independent of , in which case they are equivalent, but this should be stated.
- Section 2.1: Superscripts and subscripts are mixed: “-th player is denoted as ”. This should be not to clash with the notation for iteration counts. This should also be updated in below.
- Proof of Corollary D.2: What is In the statement and proof? This should be ? Also, the set appearing in the proof should be ?
- Top of page 15: Undefined symbol appears. Based on Fact D.4 and (A.7) this should be .
- End of proof of Lemma 2.7: The authors write: “Using (2.6) gives …, which gives (2.6)”
- Proof of corollary 2.2: Again, undefined appears, which should be ?
- Proof of corollary 2.2: The gap in the lhs, should be ?
In addition the authors should carefully proofread their manuscript, as there were several grammatical and spelling errors.
Questions for authors
Only a minor question: In (2.1), do you also assume the bounded-variance case, i.e., do you set ?
3 comments
Dear authors, thank you for your answers. I have no further questions.
Dear Reviewer JP2E,
Thank you for your follow-up message! We would be glad to discuss further if any more questions come up; to do our best to address reservations you may have about acceptance of our work.
Best regards, Authors
Thank you for the constructive feedback and for your time to carefully read our work! We really appreciated your detailed feedback.
We do not require B=0 strictly speaking, since we bound this term by compact domain assumption. With compact domains, we have uniformly bounded variance even with B>0. We use compactness in this section only for direct comparison with Ito et al. (2026). We have clarified this in Section 2.1 and included the dependence on B explicitly in the gap and regret bounds. Our results in all the other sections do not require any bounded domain or bounded variance assumptions.
We thank the reviewer for pointing out these issues. Indeed, as the reviewer stated, none of the issues affect the validity of our results, but we agree that they cause confusion. We are sorry for the confusion, we fixed all the issues in our revision.
Proof of Lemma B.3. Sorry for this typo at Lem. B.3! This was supposed to be the restatement of Lem. 3.5 where we restricted to t≥1, so 1−βt>0 throughout the proof. We agree this was confusing and it is now fixed. This bound is used for t≥1 in the proof of Lem. B.1 where the case t=0 is handled in the beginning separately.
Bottom of page 5. Indeed, you are right that the intended assumption is conditional unbiasedness, as stated in Assumption 1.2. We corrected Section 2.1 to use Et.
Section 2.1. We have corrected the notation to zi for the i-th player’s action and update the following definition of G(z).
Proof of Corollary D.2. The compact set is U, and the feasible set is Z. We have replaced B by U in the statement and prof. The remaining projC is also replaced by projZ. Top of page 15. We have replaced s by t+a+1 in the estimation (line 812).
End of proof of Lemma 2.7. We have corrected the final sentence to state that the estimate gives a uniform bound on E∣∣zt−z0∣∣2.
Proof of Corollary 2.2. Yes! We have corrected the maximum in (C.1) to be taken over u∈U.
Proof of Corollary 2.2. Yes! We have corrected the left-hand side of (C.1) to use gapU.
We really appreciated the reviewer’s detailed comments, which have helped us improve the readability of the paper. We hope that these corrections would help you reconsider your evaluation for our presentation. Can you please let us know if you have any remaining concerns?