- Rating
- 2
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 1
Read review
Paper summary
This paper studies persistent aggregation influence in federated learning when previously benign clients become malicious after a pivot at round . It proposes a method called DAIRE-FL, which maintains a decaying influence state, renews influence following accepted participation, and uses the pre-renewal state for current aggregation. Behavioral acceptance combines frozen and constrained adaptive conditional references, while attestation enforces renewal decisions. The paper derives pathwise finite-horizon bounds separating inherited influence from subsequent renewal. The paper performs experiments on CIFAR-10, CIFAR-100, FEMNIST, and Shakespeare under several adaptive attack configurations.
Critical strengths
-
The paper identifies a meaningful limitation of historical reputation. Benign participation before a behavioral pivot does not establish benign behavior afterward. Separating current aggregation authority from future renewal makes this concern explicit in practice.
-
The paper distinguishes pathwise bounds from conditional probabilistic refinements and acknowledges that enrollment does not establish identity scarcity, acceptance does not certify benign intent, and displacement bounds require bounded updates. These distinctions help prevent overly broad interpretations of the guarantees.
Critical weaknesses
-
The expected-contraction result requires malicious clients to obtain renewal infrequently enough, which is given in Eq. 13, However, the paper does not derive a bound on the malicious acceptance probability from the proposed detector DAIRE-FL or its adaptation constraints described in Section 4.2. Therefore, the result explains what happens if malicious updates are rejected sufficiently often, but does not establish when the proposed detector ensures this condition against adaptive attackers. The pathwise bound in Theorem 1 remains valid without this assumption. Nevertheless, a theoretical connection between the detector and the required acceptance rate is missing, which limits the security conclusion that can be drawn from the analysis.
-
A client receives aggregation weight only when its update is accepted, and its normalized weight cannot exceed one. Therefore, after simple calculation, it shows
Questions for authors
-
Under what conditions does the proposed detector ensure that malicious clients’ renewal rate satisfies Eq. (13)? Can this condition be established from the detector’s design and adaptation constraints, or does it require additional assumptions about the attacker?
-
Since , we immediately obtain . In which parameter regimes does Theorem 1 provide a strictly tighter bound? What additional guarantee does it offer at the paper’s default parameters?
-
The paper studies how influence changes over time, but Figure 2 mainly presents aggregate attack-success rates. Important timing details, including calibration and pivot rounds, are provided in the appendix. A timeline and a plot of client influence before and after the pivot would help readers understand what happens and why the defense works. It is a suggestion.
-
Section 5 states the mathematical conditions, but gives limited intuition about when the bound is informative. Add a clearer explanation of how inherited influence, subsequent renewals, and the horizon affect the guarantee, and what the bound establishes about the relationship between aggregation weight and actual damage in the model. In a nutshell, the practical meaning of the theorem needs more explanation.
5 comments
Responses to Reviewer NyQk’s Questions – Part 1
Question 1 (NyQk.Q1): Under what conditions does the proposed detector ensure that malicious clients' renewal rate satisfies Eq. (13)? Can this condition be established from the detector's design and adaptation constraints, or does it require additional assumptions about the attacker?
We thank the reviewer for asking whether the detector itself ensures Eq. (13). It does not. The frozen reference and constrained adaptation impose a necessary condition for acceptance, , but do not bound how often an adaptive attacker can submit updates satisfying it. A sufficient additional condition, for every pre-round history with positive selection probability and every strategy under consideration, is and . The first inequality bounds acceptance conditional on selection; the second places per-round renewal below Eq. (13)'s threshold. Under conditionally uniform -of- sampling at the controlled CIFAR defaults, a history-uniform would suffice.
The full threat model permits benign recharge by adversary-controlled identities, so neither the detector constraints nor cohort-average acceptance rates establish this condition for all allowed strategies. Selection alone caps per-round renewal at , above Eq. (13)'s approximately threshold. Expected contraction is therefore conditional, while Theorem 1 remains pathwise without this assumption (Sections 4.2–5, pages 5–6; Appendix B.8, page 25).
Question 2 (NyQk.Q2): Since each normalized weight is at most one and is zero on rejection, we immediately obtain . In which parameter regimes does Theorem 1 provide a strictly tighter bound? What additional guarantee does it offer at the paper's default parameters?
We thank the reviewer for asking when Theorem 1 improves on acceptance counting and what it guarantees at the defaults. For , Eq. (14) is strictly below exactly when . This favors stronger decay and smaller renewal increments relative to the normalization and eligibility thresholds. The condition does not hold at the controlled CIFAR defaults or listed positive-decay settings with a positive normalizer; when all normalizers are zero, realized influence is zero. For , , , , and , Eq. (14) yields . This demonstrates strictness in an unevaluated analytical regime with limited sustained eligibility (Section 5, pages 6–7; Appendix D.8, page 33).
At the unchanged CIFAR defaults, the additional guarantee comes from the separate timing-aware pathwise cap , which counts acceptances with eligible pre-renewal balance. Six consecutive acceptances yield , whereas six acceptances arranged as three spaced recharge–attack pairs yield . The three below-threshold recharge updates restore future eligibility but receive zero current weight and cannot influence detector adaptation (Section 5, page 7; Appendix B.7, pages 23–25).
We also provide an expected-weight refinement under conditionally uniform -of- sampling. Even if every selected update is accepted, the per-round renewal probability is at most . The Bernoulli envelope gives the following per-identity bounds for without assuming independent attacker renewals (Appendix B.8, page 25):
| Inherited balance | Expected count ceiling | Expected weight ceiling |
|---|---|---|
| (warm-up ceiling) |
These refinements quantify timing-dependent exclusion and expected influence under the evaluated parameters. Neither establishes an ASR bound or a uniformly tighter cap than realized acceptance counting.
Responses to Reviewer NyQk’s Questions – Part 2
Question 3 (NyQk.Q3): The paper studies how influence changes over time, but Figure 2 mainly presents aggregate attack-success rates. Important timing details, including calibration and pivot rounds, are provided in the appendix. A timeline and a plot of client influence before and after the pivot would help readers understand what happens and why the defense works. It is a suggestion.
We thank the reviewer for suggesting a timeline and a client-level view alongside the aggregate ASR results. We have retained Figure 1 for the method architecture, added a short re-entry example in Section 4.1, and added a separate balance trajectory as Figure 3 in Appendix B.7. For the illustrated controlled CIFAR-10 schedule, rounds – are calibration, rounds – are benign online participation, and the pivot occurs at round (Section 6.1, page 8; Appendix C.2, page 27).
In the constructed trajectory, the client's pivot balance is approximately . With no accepted renewals in rounds –, it falls to before round . An accepted recharge then has zero current weight and no detector-adaptation effect, but raises the next-round balance to approximately , allowing a subsequent accepted update to contribute (Appendix B.7, Figure 3, pages 23–24). The plotted quantity is the pre-round balance, which determines eligibility; it is not a measured normalized-weight or ASR curve. Normalized weight also depends on the other clients' accepted updates.
Question 4 (NyQk.Q4): Section 5 states the mathematical conditions, but gives limited intuition about when the bound is informative. Add a clearer explanation of how inherited influence, subsequent renewals, and the horizon affect the guarantee, and what the bound establishes about the relationship between aggregation weight and actual damage in the model. In a nutshell, the practical meaning of the theorem needs more explanation.
We thank the reviewer for asking for the theorem's practical meaning. The inherited balance is the state available at the pivot; it decays in subsequent rounds. An accepted renewal increases only the next balance, not the weight of its own update. Thus a long gap can make an acceptance probationary: at the controlled CIFAR defaults, a balance of falls below the threshold of after rounds without renewal. Repeated accepted renewals can restore or maintain eligibility. Increasing the horizon dissipates the inherited component, but cumulative weight need not fall when renewals continue. At these defaults, Eq. (14) does not improve on counting acceptances; the separate timing-aware cap identifies acceptances that have zero weight because the pre-renewal balance is below threshold (Section 5, page 7; Appendices B.2 and B.7, pages 17–18 and 23–25).
Theorem 1 bounds normalized aggregation weight, not attack success or the difference between complete training trajectories. If the full update norm is separately bounded by , it also bounds the sum of direct weighted update magnitudes by . Behavioral acceptance alone does not provide that norm cap, and conclusions about training outcomes require additional assumptions (Appendix B.6, pages 21–23).
Responses to Reviewer NyQk
We thank the reviewer for identifying the acceptance–contraction gap. The frozen reference and constrained adaptation do not establish Eq. (13)'s history-uniform acceptance bound against adaptive attackers. They imply only a necessary condition: since rkt≥αEθL(xkt,Sk) and τt≤τL, acceptance requires EθL(xkt,Sk)≤τL/α. The ceiling limits admissible scores, not how often attackers satisfy it, including through benign recharge.
With conditionally uniform 10-of-100 sampling at the CIFAR defaults, a history-uniform conditional-acceptance bound below 0.2880 suffices for Eq. (13)'s renewal threshold of 0.02880. Neither the detector nor cohort-average rates establish this bound; sampling alone caps per-round renewal at 0.10, insufficient for contraction below τmin. Contraction is therefore conditional, while Theorem 1 holds pathwise without an acceptance-probability assumption (Sections 4.2–5, pages 5–6; Appendix B.8, page 25).
We thank the reviewer. At the CIFAR defaults, Eq. (14) does not improve Ψk(H)≤r, where r=Rk(H). For r>0, strict improvement requires max{Zmin,τmin}(1−e−λ)>Va+bk/r. The left side is at most 3.39 at the defaults and 4.994 across listed positive-decay CIFAR settings, below Va=50.8. An unevaluated analytical example gives 17.51<r=20, although even continuous renewal approaches a balance below τmin (Section 5, pages 6–7; Appendix D.8, page 33).
At these defaults, the timing-aware pathwise cap Ψk(H)≤Gk(H)≤r counts acceptances with eligible pre-renewal balance. Starting from 100, balance falls below 30 after 25 renewal-free steps; the next acceptance restores future eligibility but has no current weight or detector-adaptation effect. Six spaced recharge–attack acceptances yield Gk=3, versus Gk=6 for six consecutive acceptances from the same pivot balance. This constrains when accepted updates contribute, not Eq. (14) uniformly or measured ASR (Section 5, page 7; Appendix B.7, pages 23–25).
We thank the reviewer for proposing a clearer Figure 1 and one-client walkthrough. We have simplified its labels while retaining enrollment, scoring, attestation, eligibility, and renewal. Section 4.1 explains re-entry, and Appendix B.7 provides the numerical Figure 3 (Section 4, page 4; Appendix B.7, pages 23–24). Benign renewals at rounds 54,64,74,84,94 yield pivot balance Bi(100)≈100.50. If submitted post-pivot attacks are rejected, no renewals occur in rounds 100–124, reducing balance to Bi(125)≈28.79<30. An accepted benign recharge at round 125 has no current weight or detector-adaptation effect but raises Bi(126)≈78.19; a later accepted attack can contribute. The constructed curve illustrates re-entry, not a measured training trace or permanent exclusion.