- Rating
- 2
- Confidence
- 4
- Informativeness
- 2
- Presentation
- 2
- Soundness
- 2
Read review
Paper summary
The paper studies the robustness of SNNs obtained through ANN-to-SNN conversion, focusing on the stability of finite-time spike-count representations under internal perturbations. The authors analyze how local state-transition probability depends on perturbation amplification and the spacing between adjacent discrete activation states. Based on this analysis, they introduce the Spike-State Stability Regularizer (SSSR), a conversion-aware regularization objective that penalizes the ratio between effective weight amplification and squared activation-state spacing. The method jointly adapts the weights of matched layers and the learnable activation thresholds during source ANN training, without requiring noise injection during source ANN training or modifications to the conversion procedure. The authors evaluate SSSR on VGG-16 and ResNet-18 using CIFAR-10 and CIFAR-100, with an additional ResNet-18 experiment on ImageNet. Comparisons with standard L1 and L2 regularization baselines, together with layerwise representation-stability analyses and ablation studies, show improved accuracy under internal Gaussian perturbations.
Critical strengths
-
The paper provides a conceptually motivated approach to improving post-conversion robustness by explicitly accounting for both effective weight-induced perturbation amplification and the spacing between discrete spike-count states. The proposed SSSR translates this analysis into a trainable regularizer that jointly adapts weights and activation thresholds, without requiring noise injection during training or modifications to the conversion procedure. This offers a potentially useful way to incorporate the properties of finite-time spiking representations into conventional ANN training.
-
The experiments on VGG-16 and ResNet-18 across CIFAR-10 and CIFAR-100 show substantial improvements over standard L1/L2 regularization baselines under strong internal Gaussian noise. The layerwise stability diagnostics and ablation studies also provide some evidence connecting the proposed regularization objective to improved representation stability.
Critical weaknesses
-
Insufficient clean accuracy and unclear accuracy–robustness trade-off. SSSR consistently achieves lower clean accuracy than the best-performing L1/L2 baseline across all four main CIFAR settings. Moreover, compared with previously reported QCFS results under the same inference settings, the gap remains substantial. The paper does not adequately explain these gaps or clarify whether the robustness gains come at the cost of clean accuracy or arise from differences in training and conversion. This raises concerns about the practical value of SSSR.
-
Limited robustness evaluation and generalization. The experiments focus almost exclusively on independent Gaussian noise injected after the input IF layer, leaving the method's effectiveness under other perturbation types unclear. Furthermore, on ImageNet, SSSR's top-1 accuracy drops from 55.80% on clean inputs to 13.75% at and 1.83% at . Although it outperforms the evaluated baselines, its absolute performance under stronger perturbations remains limited. These results provide insufficient evidence that the proposed robustness mechanism generalizes beyond the specific perturbation setting studied.
Questions for authors
-
The paper injects Gaussian noise after the input IF activation, where the per-timestep representations correspond to discrete spike events. However, additive Gaussian noise introduces continuous-valued perturbations that do not directly model errors in spike events. Would it be more realistic to evaluate spike-level errors, such as flipping 0 to 1 (spurious spikes) or 1 to 0 (missed spikes), to better assess the practical robustness of the proposed method?
-
Could the authors provide controlled comparisons with prior ANN-to-SNN conversion methods in terms of both clean accuracy and robustness under a matched perturbation protocol? Where feasible, comparisons with existing robustness-oriented SNN or conversion methods would also help clarify the contribution of SSSR relative to prior approaches.
9 comments
Response to Reviewer j9Wj (part 2 of 4)
W2 (part 1). The experiments focus almost exclusively on Gaussian noise after the input IF layer; is SSSR effective under other perturbation types?
We thank the reviewer for this suggestion. Yes: SSSR leads at the largest strength of every additional perturbation family we tested. We extended the evaluation to other noise distributions and to noise injected at other locations in the network (Table R8).
Table R8. Performances on various types of perturbations: Laplace and uniform noise of the same variance, Gaussian noise after every IF layer, and sparse impulse noise at the input IF layer or at every IF layer.
| Dataset | Method | Gaussian | Laplace | Uniform | All layers | All layers | Impulse, input layer | Impulse, all layers |
|---|---|---|---|---|---|---|---|---|
| C10 | L2-all | |||||||
| C10 | L2-wo | |||||||
| C10 | L1-wo | |||||||
| C10 | SSSR + weight decay | |||||||
| C100 | L2-all | |||||||
| C100 | L2-wo | |||||||
| C100 | L1-wo | |||||||
| C100 | SSSR + weight decay |
- SSSR leads under every noise distribution. With Laplace or uniform noise of the same variance at , SSSR with weight decay keeps % / % on CIFAR-10 and % / % on CIFAR-100, while every baseline stays below % and %.
- SSSR leads when noise enters every layer. With Gaussian noise after every IF layer at , it keeps % / %, compared with at most % / %.
- SSSR leads under sparse impulse noise. With impulses of with probability % per neuron and timestep, it keeps % / % when only the input IF layer is hit and % / % when every IF layer is hit, compared with at most % / % and % / %.
Response to Reviewer j9Wj (part 3 of 4)
W2 (part 2). On ImageNet, SSSR drops from % clean to % at and % at ; its absolute performance under strong perturbations remains limited.
We thank the reviewer for this comment. To better characterize this, we report ImageNet results across moderate and severe noise levels and additionally evaluate a deeper ResNet-34 (Table R9, Appendix C.10).
Table R9. ImageNet top-1 accuracy under moderate () and severe () noise in Appendix C.10.
| Architecture | Method | ||||
|---|---|---|---|---|---|
| ResNet-18 | L2-all | ||||
| ResNet-18 | L2-wo | ||||
| ResNet-18 | L1-wo | ||||
| ResNet-18 | SSSR | ||||
| ResNet-34 | L2-all | ||||
| ResNet-34 | L2-wo | ||||
| ResNet-34 | L1-wo | ||||
| ResNet-34 | SSSR |
- Substantial improvements under moderate noise.** With ResNet-18, SSSR achieves % and % at and , compared with % and % for the strongest conventional baseline; with ResNet-34, it achieves % and %, compared with % and %.
- Under severe noise, the baselines collapse while SSSR retains part of its accuracy. We agree that is a severe regime for 1000-class classification. Here the baselines are near chance (at most % at ), whereas SSSR keeps % (ResNet-18) and % (ResNet-34); its margin over the best baseline grows with depth from to points.
- Scientific contribution and future directions. Our primary contribution is not to fully resolve robustness under extreme noise, but to highlight an underexplored vulnerability in ANN-to-SNN conversion, identify the source-model properties governing representation stability, and introduce a principled regularizer that improves robustness without noise injection during training. On ImageNet, converted SNNs trained with L1/L2 regularization lose almost all accuracy once the internal noise reaches . This matters in practice, because converted SNNs are intended for mixed-signal or analog in-memory hardware and for split inference, where the activations after the input layer can be corrupted by readout noise or by an impaired channel. The ImageNet results demonstrate both the promise of this approach and the substantial room for further improvement. We view robustness under severe perturbations, particularly for large-scale deployment, as an important open challenge that warrants broader research efforts. The extra results on ResNet-34 have been included in Appendix C.10. Conclusion of the revised manuscript lists broader perturbation models and robustness in large-scale deployment settings as future work.
Response to Reviewer j9Wj (part 1 of 4)
We thank the reviewer for the detailed review and for recognizing the conceptual motivation of SSSR and its gains over L1/L2 regularization.
W1. SSSR has lower clean accuracy than the best L1/L2 baseline and a gap to QCFS; do the robustness gains come at the cost of clean accuracy or from differences in training and conversion?
We thank the reviewer for this constructive comment. We have done further investigation and confirmed that the gap in our reported QCFS results to the QCFS paper came from our training hyperparameters. With the QCFS hyperparameters, we have successfully repeated their results, and confirmed that the robustness gain of our proposed SSSR remains unchanged. Moreover, SSSR is only 0.3 / 1.0 points below L2-wo in clean accuracy. Detailed findings are as below.
(1) Source of the gap to QCFS. The gap came from our choice of training hyperparameters. Our CIFAR VGG-16 setup used Dropout(0.5) after every layer and a batch size of 128, whereas the official QCFS code uses no Dropout and a batch size of 300. We re-ran every method with the QCFS hyperparameters and report their performances below.
Table R6. VGG-16 retrained with the hyperparameters of the official QCFS code (mean ± std); the first row is L2-all with our hyperparameters used in the paper at the time of submission.
(2) The clean-accuracy cost of SSSR is small. With weight decay of 10−4 on the weights, we are able to reduce the gap of SSSR and baseline on clean-accuracy, using both VGG-16 and ResNet-18.
Table R7. The same comparison on ResNet-18 (mean ± std).
We have stated this trade-off explicitly in Section 4 and Appendix C.8 of the revised manuscript.