- Rating
- 2(+1)
- Confidence
- 3
- Informativeness
- 2(+1)
- Presentation
- 2
- Soundness
- 2(+1)
Read review
Paper summary
This paper studies state-dependent residual transmission in SNNs by holding the merged input M fixed and replacing the retained membrane state R. It derives an exact threshold-crossing condition for additive LIF and PLIF neurons and examines how state pairing affects switching prevalence. Experiments on SpikeDF show that substantial spike switching can coexist with nearly unchanged mean output, while controlled spike substitution reveals site-dependent downstream sensitivity. Continuous-merge FF++ and DVS128 Gesture experiments further show that switching concentrates near the firing boundary. Separate recursive state interventions assess how retained state affects task performance, distinguishing local switching susceptibility from downstream sensitivity and trajectory-level dependence.
Critical strengths
Strengths
-
The experimental design is grounded, and the coverage of experimental settings is broad. Thus, I think the experimental results are reliable.
-
The authors provide mathematical and theoretical explanations about how status exchanges of LIF neurons affect the neurons' spiking patterns.
Critical weaknesses
Weaknesses
-
In my opinion, the problem tackled by the authors is not significant. The problem is somewhat detailed, but I cannot assure that this problem significantly degrades the learning performance of residual SNNs. The authors fail to clarify how the tackled problem causes further limitations in residual SNNs.
-
The authors measure how often the output spike of a post-add LIF neuron changes when its retained membrane state is replaced while the residual input remains fixed. However, I am not sure that these changes in spiking patterns cause negative effects. Changes in spiking patterns of LIF neurons are normal.
-
The authors only use LIF-based residual SNNs. I think they should extend their analysis using various neuron models. For example, residual SNNs can be implemented with IF neurons [1].
-
The authors use video forgery detection tasks to evaluate the SNNs' performance. Although forgery detection is a reasonable temporal task, the paper does not sufficiently justify why this application is particularly relevant to the proposed analysis.
[1] W. Fang, Z. Yu, Y. Chen, T. Huang, T. Masquelier, and Y. Tian, "Deep Residual Learning in Spiking Neural Networks," in Advances in Neural Information Processing Systems (NeurIPS), vol. 34, pp. 21056–21069, 2021.
Questions for authors
Questions
-
First of all, does the observed phenomenon really cause negative effects on SNNs?
-
Although the authors state that spiking activities are affected by residual connections, they do not show experimental results with changing the configurations of residual connections in the paper. Why didn't you change these configurations? I think it might be important to support the authors' claim.
-
Why did you use temporal residual cues and hard pseudo-events in SpikeDF? Are they widely used pre-processing techniques?
12 comments
Q1 : Does the phenomenon have a functional effect? (W1, W2)
W1 : Retained history and task performance
We thank the reviewer for asking about consequences for the trained detector. Retaining the history associated with the current video has a substantial functional role: when we pair a video with another video’s history, FF++ validation AUC falls from 90.78 ± 0.16% to 75.90 ± 0.68% (paired change percentage points), although late activity changes only from 4.80 ± 0.14% to 4.48 ± 0.37%. The contrast shows that the detector depends on the history accumulated with the current input, even when overall late firing activity changes relatively little.
Stage-local recurrent interventions further reveal that this dependence varies with the parts of the network affected:
| Retained-history intervention | FF++ AUC (%) | Paired change (pp) | Late activity (%) |
|---|---|---|---|
| Self history | 90.78 ± 0.16 | — | 4.80 ± 0.14 |
| Clear Stage 1 | 49.25 ± 5.87 | −41.53 ± 5.82 | 2.75 ± 0.33 |
| Clear Stage 2 | 57.79 ± 2.84 | −33.00 ± 3.00 | 2.54 ± 0.52 |
| Clear Stage 3 | 90.59 ± 0.11 | −0.19 ± 0.18 | 1.63 ± 0.14 |
| Clear Fusion | 90.51 ± 0.08 | −0.27 ± 0.20 | 3.04 ± 0.11 |
| Clear Stage 3 and Fusion | 45.90 ± 0.59 | −44.88 ± 0.50 | 0.00 ± 0.00 |
Clearing Stage 3 or Fusion separately changes AUC by only 0.19 or 0.27 points, whereas clearing both decreases it by 44.88 points and leaves zero late activity. This non-additive task response shows why the functional role of retained history cannot be inferred by adding the effects of separate stage interventions.
W2 : Spike identity, mean activity, and downstream sensitivity
The fixed-input experiments explain a complementary part of this behavior. At S3.1, empirical random-donor switching is 13.61%, versus 49.76% for the independent-marginal reference within merged-current bins, demonstrating the influence of observed state pairing. At S3.0, the nearest-donor experiment changes 13.33% of selected outputs while the net mean-activity change is only −0.04 points, because switches in opposite directions largely cancel.
The effects are visible downstream as well: propagating all identified S1.0 output substitutions through the recorded downstream computation changes the predicted label of 19.53 ± 1.56% of evaluated videos. With the same budget of 100 identified substitutions at each site, the relationship between local switching and detector-score response is:
| Site | Nearest-donor switching (%) | Mean at |
|---|---|---|
| S1.0 | 33.46 ± 0.83 | 0.1677 ± 0.0037 |
| S3.0 | 13.33 ± 1.69 | 0.0658 ± 0.0044 |
| S3.2 | 15.26 ± 0.75 | 0.0424 ± 0.0049 |
S3.2 switches more often than S3.0, yet produces a smaller detector-score change under the same substitution budget. By linking the observed state-pairing statistics to spike substitutions and detector responses, these measurements explain why neither local switching prevalence nor aggregate firing activity is sufficient on its own to characterize downstream sensitivity. The separate recurrent-history interventions then show how changes to the input–history relationship affect task performance over the full sequence.
Q2 : Why compare these residual configurations? (W3)
Residual depth and transmission rules
We appreciate the suggestion to vary residual configurations, which helps clarify the scope of our experimental comparisons. We vary residual depth and the rule applied after the branch–shortcut merge, while holding the wiring pattern consistent so that a topology change is not introduced alongside the post-add rule. In the separate fixed-input analysis, the realized merge itself is fixed; the independently trained rule variants share the wiring but need not produce identical merged inputs.
The four- and six-block FF++ comparisons include Static Supra-Unit, Fixed-Stateful LIF, learned RD-LIF, and Binary-OR, with an additional SEW-ADD-style operator that preserves multivalued residual outputs:
| Post-add rule | 4-block FF++ AUC (%) | 6-block FF++ AUC (%) | 6-block late active-state density (%) |
|---|---|---|---|
| Static Supra-Unit | 82.23 ± 1.65 | 73.97 ± 1.71 | 0.089 ± 0.023 |
| Fixed-Stateful LIF | 90.95 ± 0.23 | 91.31 ± 0.65 | 4.92 ± 0.75 |
| Learned RD-LIF | 91.08 ± 0.37 | 90.76 ± 0.20 | 4.80 ± 0.14 |
| Binary-OR | 92.27 ± 0.61 | 92.52 ± 0.89 | 85.31 ± 0.46 |
| SEW-ADD-style | 92.44 ± 0.58 | 92.77 ± 0.46 | 75.18 ± 0.25 |
In the six-block setting, the alternatives thus yield markedly different activity regimes and AUCs. We also retrained the models after removing the auxiliary continuous residual-statistics pathway:
| Post-add rule, no auxiliary pathway | FF++ AUC (%) | Late activity (%) |
|---|---|---|
| Static Supra-Unit | 75.78 ± 1.50 | 0.07 ± 0.03 |
| Fixed-Stateful LIF | 90.84 ± 0.10 | 5.17 ± 0.32 |
| Learned RD-LIF | 90.47 ± 0.56 | 4.95 ± 0.15 |
Across three paired checkpoints, Fixed-Stateful LIF and learned RD-LIF exceed Static Supra-Unit by 15.06 ± 1.57 and 14.69 ± 1.24 AUC percentage points, respectively, with positive differences in every checkpoint. These results establish the importance of the evaluated post-add transmission choices within matched residual wiring, while the fixed-input analysis isolates the effect of membrane history at a realized merge.
W3 : IF applicability and trained neuron-model coverage
Our independent trained-network tests include learned RD-LIF in SpikeDF, standard LIF with continuous FF++ merges, and PLIF with native DVS128 Gesture events. For an additive IF update without leak, the normalized pre-threshold state can be expressed as , so two retained states switch the output at a fixed input when
The crossing condition therefore applies to additive IF analytically.
We appreciate the reviewer’s careful consideration of the functional role of retained history in residual SNNs. Our fixed-input analysis connects the neuron-level switching condition to what is observed in trained networks: the switch rate depends on how membrane states are paired, while the effects of those switches cannot be inferred from mean firing activity alone. At S3.1, random-donor replacement produces 13.61% switching, compared with 49.76% under the independent-marginal reference within merged-current bins. At S3.0, 13.33% of selected outputs switch even though mean activity changes by only −0.04 percentage points, and equal-count substitutions at different residual sites lead to different detector-score responses. Full recurrent-history interventions further show how the history associated with an input affects task performance. We address the reviewer’s specific questions below.