- Rating
- 3(+1)
- Confidence
- 2
- Informativeness
- 3
- Presentation
- 3
- Soundness
- 2
Read review
Paper summary
This paper proposes MRD-SNN, which augments spiking neurons with dendritic memory branches operating at different update rates and memory horizons. Each branch maintains a Legendre memory and supplies a cached readout between updates. The method is evaluated on three neural decoding tasks, SHD, and sequential MNIST. The authors also study memory capacity, branch combinations, and the accuracy-efficiency trade-off of reducing memory update frequency.
Critical strengths
-
The paper provides a concrete way to combine temporal memory with sparse updates. Separating the memory horizon from its update frequency is useful for controlling computational cost. The matched-horizon comparison reports 1.55-1.78x speedups with accuracy differences of at most 0.74 percentage points, which supports the practical value of reducing refresh frequency.
-
The empirical results are promising.
Critical weaknesses
-
It is not convinced that the accuracy gains come specifically from using different update rates. In Table 4, changing the update interval at fixed also changes the physical memory horizon. When the horizons are matched in Table 9, full-rate and multi-rate updates achieve similar accuracy. These results make a stronger case for reducing computation than for improving accuracy through rate diversity.
-
The shared-versus-independent memory comparison in Table 5 also changes the memory budget substantially: moving from to increases the number of dendritic memory-state variables 128-fold. It is therefore difficult to tell how much of the improvement comes from giving each neuron its own memory and how much comes from simply having more memory capacity.
Questions for authors
-
Could you repeat the mixed-rate versus same-rate comparison on SHD and sMNIST while keeping the same set of physical memory horizons? Does the advantage reported in Table 4 remain under this control?
-
Do neuron-specific memories still outperform shared memories when the total number of memory-state variables is matched?
Q1:
We thank the reviewer for suggesting this important control. We added comparisons that keep the same pair of physical memory horizons while changing only the refresh intervals, allowing us to examine the effect of update scheduling separately.
The horizons are 0.112 and 1.12 seconds on SHD, and 98 and 980 input steps on sMNIST. Each branch has q=10; parameter counts, connection masks and trainable initializations are identical. We adjust θ with n to keep H=nθΔt fixed.
Results use five folds and three seeds per fold, with validation-selected checkpoints. Mixed and high-rate refresh achieve nearly the same accuracy: mixed refresh is lower by approximately 0.7 and 0.4 percentage points, while reducing nominal branch-update counts by approximately 45%.
Our design considers both memory specialization and computational cost. A high-rate branch can obtain a long horizon through a larger θ, but frequent refresh does not necessarily imply a rapid response to new input. At fixed memory order, when ∥A0∥/θ≪1, the branch update can be approximated as
mj≈mj−1+θ1(A0mj−1+B0xj)
Here, mj is the memory state, xj is the current input, and A0 and B0 are the unscaled Legendre matrices. A larger θ weakens both the write from new input and the change to the previous state at each update, making the branch a slowly evolving summary of history. A high-rate, long-horizon memory therefore does not automatically provide both recent detail and distant context.
Appendix G further illustrates the risk of performance failure for a single high-rate, long-horizon branch on tasks requiring very long retention together with recent timing information. On the 8192-step XOR task, increasing the single branch to q=20 yields only 57.75 ± 12.21% accuracy, whereas a high-rate short-memory branch paired with a low-rate long-memory branch reaches 99.98 ± 0.03%.
We therefore let different branches specialize in recent detail and distant context, and reduce the long branch’s refresh rate to lower update costs. When the required memory span is shorter, two high-rate branches with different θ values also perform well; multirate scheduling can retain similar accuracy while reducing update overhead.
Q2:
We appreciate the reviewer’s observation that the original comparison also changes memory capacity. We therefore added comparisons on all three neural decoding datasets that keep the total number of dendritic memory states fixed while varying the degree of memory sharing.
We retain the four-branch architecture, update intervals, nominal memory horizons and training protocol of Table 5, and adjust the memory order q as the number of memory groups G changes. Every configuration has the same total number of dendritic memory states:
S=2×4×Gq=10,240,Gq=1,280
Here, G=1 shares memory across the entire layer, whereas G=128 gives each neuron its own memory.
Accuracy (%) is reported as the mean ± sample SD across five folds, after averaging three seeds within each fold.
Neuron-specific memory achieves the highest mean accuracy on all three datasets under the matched state budget, exceeding the best shared-memory configuration by 1.08, 0.55, and 4.01 percentage points, respectively. These results support the value of maintaining separate temporal contexts for individual neurons even when the total number of memory states is fixed.