- Rating
- 3
- Confidence
- 2
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
This paper studies the design of redundant block frames for transmitting a latent representation over a block-erasure channel when the downstream task is multi-class classification. The key observation is that, with erasure locations known and least-squares decoding, every full-column-rank survivor frame recovers exactly the same latent at zero noise, so the remaining design freedom can be redirected from reconstruction criteria toward reducing decision uncertainty under bounded observation noise. Based on these, the paper makes three contributions: (1) derives an exact worst-case pairwise score-uncertainty certificate and the exact per-sample all-survivor certified radius, (2) solves the rank-relaxed trace-constrained allocation in closed form, proves minimax optimality of an anisotropic allocation via polar factorization, and proves that the rank constraint makes the relaxation unattainable, (3) realizes TACBF, combining the closed-form allocation with a learned Parseval factor trained by a certificate loss. On seven datasets with five paired seeds at (m, q) = (8, 2), TACBF improves certified-accuracy AUC over a matched fixed-factor control on all datasets (all 35 paired seed differences positive, significant after Holm correction), preserves exact recovery, width, trace, and the LS decoder, and changes clean accuracy by at most 0.25 pp, at the cost of a 3.4–65.2% increase in expected LS reconstruction error.
Critical strengths
- The paper proposes a novel formulation built on a full certificate-to-design theory. It innovatively repurposes the design freedom from noiseless exact recovery for optimizing decision geometry, departing from prior frame design focused on reconstruction metrics. Beyond the robustness certificate, it derives closed-form relaxed optimum, minimax allocation and proves the rank constraint gap, forming a complete theoretical storyline supporting this positive evaluation.
- The experiments adopt rigorous methodology. It uses seed-paired baselines to isolate the learned factor, enumerates all survivor sets for certification, applies proper statistical testing, and transparently reports negative results. Such strict controls render empirical claims reliable.
- Empirical improvements are consistent across diverse settings with tiny deployment overhead. The method works on seven datasets of three modalities, with consistent positive gains over all seed pairs. It introduces marginal training and memory cost while preserving inference speed and clean accuracy.
Critical weaknesses
- A restrictive operating model and purely synthetic evaluation limit the demonstrated significance. The certificate relies on known erasure locations, deterministic ℓ2-bounded observation noise, a linear classification head, and unregularized least-squares decoding, which the authors acknowledge in Section 8. The experiments add noise in feature space instead of simulating a real communication channel. In addition, no existing baseline accounts for the exact-recovery constraint; block dropout serves as the closest comparison but does not optimize for the certificate objective. The core mathematical component of Theorem 1, the exact worst-case score perturbation of a linear classifier over an ℓ2 ball under a linear transformation, draws from prior classical work such as Hein & Andriushchenko (2017). The real novelty of this paper comes from combining this formulation with least-squares decoding, optimizing over the worst case across survivor sets, and developing the accompanying design theory. The work is technically solid. Even so, in its current form, the paper reads primarily as a strong contribution to frame theory rather than a practical robust communication solution.
- Certified-accuracy AUC is evaluated over author-selected parameter grids using inconsistent metrics and ranges, which deviates from standard community protocols of fixed-radius certified accuracy or full certification curves. Several observed improvements are marginal relative to baseline values, and gains on the Banking77 dataset only appear at small radii, concentrating where the baseline performs poorly. The primary experiments adopt the setting (m,q)=(8,2); tests with m=16 are limited to an ablation search, and direct training for q=3 is only conducted on one dataset with three seeds. Besides, exhaustive certification incurs combinatorial computational cost scaling with (m¦q). 3.The certified-classification gains come at a substantial, quantified reconstruction cost that weakens the “exact latent recovery” value proposition under noise. Expected squared LS reconstruction error rises 3.4–65.2% (e.g., PCam 12.59 vs. 7.63), with task-orthogonal directions de-amplified by design, so any downstream use of the latent beyond the fixed linear head inherits noisier recovery. The tradeoff is characterized exactly (Eqs. 10, 19), but no practical interpolation is studied (e.g., sweeping μ).
Questions for authors
Q1. Does the certificate extend to input-space perturbations or stochastic (e.g., Gaussian) channel noise? If not, please state this scope explicitly, since “decision uncertainty” may be read as input-adversarial robustness. Q2. What happens with unknown or imperfectly estimated erasure locations — is the all-survivor certificate still valid, or does a wrong mask break exact recovery and the radius? Q3. How much certified robustness does the exact-recovery/Parseval/floor machinery cost? A certificate-aware baseline without these constraints, under the same channel and budget, would price the main constraint.
5 comments
Response to Reviewer U2ZL: Weakness 2
W2. AUC protocol, radius-dependent gains, redundancy and certification cost.
Certified-accuracy AUC is evaluated over author-selected parameter grids using inconsistent metrics and ranges, which deviates from standard community protocols of fixed-radius certified accuracy or full certification curves. Several observed improvements are marginal relative to baseline values, and gains on the Banking77 dataset only appear at small radii, concentrating where the baseline performs poorly.
We appreciate your request for full curves and fixed-radius results; these show where the gains occur. Table 1 (p. 7) reports fixed-radius accuracy; Figures 2 and 5 (pp. 8 and 31) show full curves alongside grid and exact continuous AUC. The original fixed-radius results are:
| Dataset | Radius | Control (%) | TACBF (%) |
|---|---|---|---|
| CIFAR-10 | 84.92 | 87.02 | |
| STL-10 | 14.81 | 47.33 | |
| EuroSAT RGB | 80.52 | 90.54 | |
| CIFAR-100 | 19.60 | 51.40 | |
| PCam | 65.28 | 69.48 | |
| UCI HAR | 61.26 | 69.30 | |
| Banking77 | 0.00 | 0.10 |
For Banking77, the following points show gains where the control still certifies a substantial fraction of examples:
| Radius | Control (%) | TACBF (%) | Gain (pp) |
|---|---|---|---|
| 75.85 | 79.02 | +3.17 | |
| 46.30 | 57.32 | +11.03 | |
| 8.46 | 23.87 | +15.41 |
The control still certifies 75.85% at , and the largest gain among these points is 15.41 pp at . The Banking77 curve in Figure 5 (p. 31) shows this range alongside the tail at , where both methods are close to zero. Exact continuous AUC improves in all 35 paired runs; Tables 12–13 (p. 32) report the paired intervals and Holm-adjusted tests for the seven datasets:
| Dataset | ΔAUC | Paired 95% CI | |
|---|---|---|---|
| CIFAR-10 | +0.0078 | [0.0030, 0.0126] | 0.0212 |
| STL-10 | +0.0797 | [0.0623, 0.0972] | 0.0011 |
| EuroSAT RGB | +0.0153 | [0.0127, 0.0178] | 0.0004 |
| CIFAR-100 | +0.0357 | [0.0274, 0.0440] | 0.0011 |
| PCam | +0.0562 | [0.0400, 0.0724] | 0.0019 |
| UCI HAR | +0.0190 | [0.0074, 0.0307] | 0.0212 |
| Banking77 | +0.0308 | [0.0293, 0.0324] |
Section 6 (pp. 6–7) defines the axes: absolute for CIFAR-10/STL-10/EuroSAT, and for the other datasets, where . Both methods use the same fixed control's training energy as reference, and the ranges were fixed using training/development data. Exact integration removes dependence on grid spacing. With for a clean error, normalized continuous AUC on is
Besides, exhaustive certification incurs combinatorial computational cost
In the main setting there are masks, and the worst-survivor coefficient for each class pair is cached across test examples. For larger settings, branch-and-bound supplies valid radius intervals at a search budget and is exact on completion; worst-case growth remains combinatorial. The separate search diagnostic takes 10.4 s for enumeration; beam width eight visits 295 nodes per pair on average, with empirical mean relative gap . Appendix D (pp. 28–29) and Figure 4 (p. 29) report the search results.
For configuration coverage, frozen models trained at are evaluated at on all seven datasets. The common- AUC gain rises from 0.0040 at to 0.0723 at , and the gain exceeds in all 35 paired runs. A separate STL-10 study with three paired seeds trains directly at and gives mean grid AUC 0.3728 versus 0.1781 for the control. The experiment evaluates search scalability. Appendix F.1 (p. 34) reports transfer and direct training separately, with Figure 7 showing transfer. Appendix G.4 (Table 22 and Figure 12, p. 41) reports sensitivity to training radius and survivor sampling.
Response to Reviewer U2ZL: W3 & Q3
W3. Reconstruction tradeoff.
The certified-classification gains come at a substantial, quantified reconstruction cost [...] no practical interpolation is studied (e.g., sweeping μ).
Keeping representation, head, , width and trace fixed, we swept the floor and squared allocation (: isotropic; : analytic). All nine settings meet their floors; five seeds per dataset.
Working points: AUC / full unit-sphere MSE.
| CIFAR-10 | PCam | Banking77 | |
|---|---|---|---|
| isotropic | 0.308211 / 1.2607 | 0.070365 / 1.2339 | 0.202310 / 1.2059 |
| 0.827118 / 2.1335 | 0.537281 / 2.2337 | 0.231653 / 1.4565 | |
| default | 0.920851 / 10.2571 | 0.690974 / 12.5940 | 0.257676 / 5.9745 |
| 0.922939 / 41.0097 | 0.698943 / 50.3756 | 0.261342 / 22.5822 | |
| 0.884185 / 2.5722 | 0.622815 / 3.1486 | 0.242456 / 1.8758 |
At , full MSE falls 49.96%, 50.00%, 46.07% versus default; AUC losses are .00383, .01318, .00497 (dataset order above). Mean per-seed task MSE rises 13.73%, 12.02%, 3.97%, while orthogonal MSE is about halved.
Let be the task and orthogonal traces of , and the corresponding eigenvalues of . For Gaussian noise,
For unit-sphere noise, divide by the observation dimension. Increasing orthogonal allocation reduces reconstruction error at the expense of task protection. Table 3 (p. 9) and Appendix G.5 (pp. 42–44) report these paths.
Q3. Joint constraint cost and W1's baseline concern.
W1 asks:
[...] block dropout serves as the closest comparison but does not optimize for the certificate objective.
Q3 asks:
How much certified robustness does the exact-recovery/Parseval/floor machinery cost? A certificate-aware baseline without these constraints, under the same channel and budget, would price the main constraint.
The original fixed-factor control retains exact zero-noise recovery, analytic allocation and the certificate objective; block dropout uses survivor cross-entropy. The new full-width refit matches the objective, frozen representation, head, width, trace and schedule, with only a global Frobenius energy constraint. It imposes no Parseval structure, prescribed allocation, survivor floor or full-rank constraint, and uses its actual pseudoinverse.
Full-width comparison: AUC / full unit-sphere MSE; effective paired n.
| Dataset | TACBF | Full width | n |
|---|---|---|---|
| Original trace | |||
| CIFAR-10 | 0.691807 / 3.7248 | 0.010189 / 10599.3083 | 2 |
| PCam | 0.677964 / 11.8392 | 0.711418 / 90.0242 | 3 |
| Banking77 | 0.252630 / 5.6556 | 0.253261 / 1.4918 | 3 |
| Matched mean energy | |||
| CIFAR-10 | 0.433178 / 35.0231 | 0.029746 / 30203.5456 | 2 |
| PCam | 0.667403 / 1049.2702 | 0.695435 / 8761.1809 | 3 |
| Banking77 | 0.248185 / 9.1481 | 0.229367 / 2.8390 | 3 |
One CIFAR-10 TACBF fit fails the floor (n=2). At original trace, PCam full width gains .03345 AUC with 7.60 times TACBF's MSE. At matched energy, Banking77 TACBF gains .01882 AUC and 7.22 pp Gaussian accuracy at -10 dB (59.60% vs. 52.38%; paired 95% CI [6.66, 7.78], n=3), with 3.22 times full width's MSE. CIFAR-10 full-width fits are severely ill-conditioned.
All nine full-width fits were full rank on all 28 masks and recovered exactly at zero noise: observed outcomes, not training guarantees.
Any full-column-rank admits , , so Parseval factorization preserves capacity. This comparison measures the joint effect of allocation, floor and training.
After energy matching, all eight TACBF fits retain full rank but fall below the original .01 floor. Appendix G.6 (pp. 45–46), Tables 27–28, reports rank, conditioning, floors and all five Gaussian SNRs.
We especially appreciate your suggestion to study the reconstruction tradeoff in practice. It helped us show how the floor and allocation control the balance between classification robustness and latent recovery. We hope these additions address your concerns and would welcome your thoughts on the new comparisons.
Response to Reviewer U2ZL: W1 & Q1 & Q2
Thank you for your thoughtful review and for recognizing the theoretical development and experimental controls. Your questions led us to add Gaussian tests and a comparison with freely parameterized frames. We provide detailed responses to each of your concerns below.
W1. Operating model and contribution.
We study classification with complete zero-noise latent recovery under known block erasures and LS decoding. A camera could send latent features in identified packets for industrial inspection (e.g., classifying components as defective or intact). The receiver could then make an immediate decision and retain the recovered latent for later diagnosis. All admissible frames recover the same latent at zero noise, but amplify noise differently across directions; design protects directions relevant to the decision. We evaluate on real image, activity and intent benchmarks, with noise and erasures introduced in the latent channel.
The linear support function calculation is classical; Section 2 (p. 2) cites related certification work, including Hein and Andriushchenko (2017). The contribution is a frame design rule: the exact radius over all survivor sets identifies decision directions to protect; relaxed allocation specifies an ideal task/orthogonal energy split; the strict block-rank gap explains why finite blocks cannot realize the corresponding survivor geometry. Allocation and geometry optimization are complementary. Figure 3 (p. 10) illustrates the geometry's effect.
Q1. Input perturbations and Gaussian noise.
The deterministic certificate concerns additive observation noise on surviving received blocks. For ηC∼N(0,σ2I), GC=AC⊤AC, v=wy−wk and clean margin myk>0, LS gives the exact pairwise crossing probability
Pr(myk≤0∣C)=Φ!−σv⊤GC−1vmyk.
Here Φ is the standard normal CDF. We measure multiclass accuracy directly, accounting for correlated competitors. At matched training mean transmitted energy and -10 dB, TACBF/fixed-factor control accuracies are 93.17/92.80% (CIFAR-10), 83.47/83.44% (PCam) and 60.39/55.13% (Banking77). Banking77 gains 5.26 pp, paired 95% CI [5.01, 5.50]; CIFAR-10 gains .37 pp, and PCam is mixed across settings. Appendix G.3 (pp. 38–40), Tables 20–21 (p. 39) and Figures 10–11 (pp. 39–40) report the Gaussian tests and reconstruction cost.
For an input perturbation ξ, noiseless LS returns f(x+ξ). An input certificate therefore requires a bound on the encoder's response, such as a Lipschitz bound for f; the present channel certificate alone does not supply it. Sections 3 and 8 (pp. 3 and 10) state this scope.
Q2. Erasure-mask identities.
A wrong mask can break exact recovery and the radius because LS uses the supplied block identities. If Lassumed is the decoder and Atrue the actual survivor operator,
s−s=(LassumedAtrue−I)s+Lassumedη.
The first term is a zero-noise bias. A scalar encoded with coefficients 1 and 2 is recovered from either correctly identified packet; treating the second as the first returns 2s. The all-survivor guarantee covers every admissible set with its true block identities. For an unknown mask, the block identities must first be recovered before using the present decoder and certificate. Section 3 (p. 3), Eq. (3), gives the decomposition.