Reducing ANN-SNN Conversion Error via Residual Membrane Potential Alignment

Zirui Chen; Zihan Huang; Tong Bu; Jianhao Ding; Yiting Dong; Zhaofei Yu

cs.NEAug 14, 2026

Reducing ANN-SNN Conversion Error via Residual Membrane Potential Alignment

Authors: Zirui Chen, Zihan Huang, Tong Bu, Jianhao Ding, Yiting Dong, Zhaofei Yu

Organizations: Institute for Artificial Intelligence, Peking University, and also with Beijing Key Laboratory of Brain-inspired Spiking Large Models, the School of Computer Science, Peking University, Beijing, China · School of Computer Science, Peking University, and also with the National Key Laboratory for Multimedia Information Processing, Peking University, Beijing, China

Abstract

Spiking Neural Networks (SNNs) serve as core architectures for neuromorphic computing thanks to event-driven operation and ultra-low power consumption. Direct SNN training is hindered by non-differentiable spikes that induce vanishing gradients and unstable optimization. ANN-SNN conversion circumvents such issues by reusing well-trained ANN weights for low-latency, energy-efficient inference. Nevertheless, existing conversion schemes suffer from severe accuracy drops at small timesteps, large inference delays and cumulative quantization errors, even with marginal performance loss at large $T$ . To address these limitations, we first analyze flaws of conventional conversion pipelines from residual membrane potential statistics and propose a novel conversion strategy combining dynamic initial potential tuning and feature enhancement. We then introduce a regularization loss $\mathcal{L}_{\mathrm{RMPD}}$ to adapt initial potential of IF neurons and mitigate systematic truncation bias from boundary aggregation. A dedicated SCR-Conv2d competitive refinement layer with grouped convolution is further built to sharpen feature discrimination, eliminate redundant spikes and stabilize encoding under tiny time windows. Integrated with the state-of-the-art QCFS baseline, our approach delivers consistent low-latency performance gains and generalizes to ReLU CNNs, ANN Transformers, and multi-threshold SNN variants. Evaluations on CIFAR-10, CIFAR-100 and ImageNet verify prominent accuracy improvements at $T=2,4,8$ , with negligible extra computation overhead. This work offers an effective conversion paradigm to facilitate real-world SNN deployment on neuromorphic chips.

Figures & tables

Fig. 1: Distribution of residual membrane potential. The subplots demonstrate RMPD under distinct input distributions. Case (a) is dominated by the interval

[\frac{\theta}{L},\frac{2\theta}{L}]

, while case (b) receives equal contributions from

[0,\frac{\theta}{L}]

and

[\frac{\theta}{L},\frac{2\theta}{L}]

. Correspondingly, the RMP distribution in (a) exhibits near-Gaussian behavior, whereas the distribution in (b) severely departs from Gaussianity. Figure (c),(d) correspond to figure (a),(b) with smaller variance. Comparing case (c),(d) with (a),(b), we can tell that reducing variance of neuron input amplify the error-reduction effect.

Fig. 2: Fine-tuning of the initial membrane potential

m_{\mathrm{init}}

. The gradient step for updating

m_{\mathrm{init}}

is proportional to the offset of the target interval. After each adjustment, the resulting RMPD becomes closer to a normal distribution, and the corresponding spike activations cluster more tightly around the characteristic firing frequency of each neuron.

Fig. 3: RMPD shapes before and after adjustment. Parameter QQ quantifies the similarity to Gaussian distribution, and it can be observed that the regularized RMPD is closer to Gaussian distribution after adjustment.

Fig. 4: Fine-tuning of initial membrane potential with shift compensation

Fig. 5: Architecture of SCR-Conv2d. The SCR-Conv2d layer is inserted after the standard convolutional layer. The outputs of the IF layer are transformed into refinement signal magnitudes via the nonlinear function $F_{\mathrm{nl}}$ . An additional IF layer is appended after SCR-Conv2d to maintain spike transmission. This figure illustrates the spike-driven information propagation pipeline of the converted SNN within VGG-16.

Method	Net.	Param. (M)	ANN	2	4	8	16	32	64
CIFAR-10
QCFS [ 41 ]	VGG16	138.36	95.78	90.67	94.21	95.40	95.73	95.74	95.83
SNM [ 44 ]	VGG16	138.36	95.72	90.55	94.54	95.56	95.78	95.77	95.82
SRP( $\tau$ =4) [ 18 ]	VGG16	138.36	95.62	$-$	95.43	95.55	95.58	95.57	95.60
NQQCFS [ 37 ]	VGG16	138.36	95.21	91.93	94.80	95.48	95.70	95.79	95.72
SCR-SRMP	VGG16	138.36	95.65	92.84	94.80	95.66	95.78	95.80	95.86

TABLE I: Comparison with state-of-the-art methods on the CIFAR-10,CIFAR-100 and ImageNet dataset

Arch.	Method	Param. (M)	Original (ANN)	SNN
Arch.	Method	Param. (M)	Original (ANN)	$T=2$	$T=4$
CIFAR-10
ViT-S/16	ECMT [ 69 ]	22	98.33	24.81	92.78
ViT-S/16	SRMP	22	98.33	26.62	94.31
ViT-B/16	ECMT [ 69 ]	86	98.73	85.67	96.76
ViT-B/16	SRMP	86	98.73	87.17	97.03

TABLE II: Comparison results on Transformer

Method	Conversion Type	Arch.	Param. (M)	T	$T_{ratio}$	Acc(%)
QCFS [ 41 ]	ANN-to-SNN	ResNet-34	21.8	64	$-$	72.35
QCFS [ 41 ]	ANN-to-SNN	VGG-16	138	64	$-$	72.85
SRP [ 18 ]	ANN-to-SNN	ResNet-34	21.8	4(64)	0.0625	66.71(68.61)
SRP [ 18 ]	ANN-to-SNN	VGG-16	138	4(64)	0.0625	66.46(69.43)
MST [ 73 ]	Transformer-to-SNN	Swin-T(BN)	28.5	128(512)	0.25	77.88(78.51)
STA [ 74 ]	Transformer-to-SNN	ViT-B/32	86	32(256)	0.125	78.72(82.79)

TABLE III: Comparison with previous works on ImageNet1k dataset

Panel of Figure 9 — Fig. 6: Variations of $Loss$ and $\mathcal{L}_{\mathrm{RMPD}}$ during fine-tuning on CIFAR-10, CIFAR-100 and ImageNet. $\mathcal{L}_{\mathrm{RMPD}}$ steadily decline, while $Loss$ fluctuate mildly or rise slightly.

Fig. 7: Curves of

\mathcal{L}_{\mathrm{RMPD}}

induction in ViT-S/16, ViT-B/16 and ViT-L/16 model on CIFAR-10, CIFAR-100, ImageNet. ALL the curves decline steadily and flatten out over time.

Method	Net.	ANN	2	4	8	16
CIFAR-10
QCFS [ 41 ]	VGG16	95.78	90.67	94.21	95.40	95.73
SRMP	VGG16	95.60	91.49	94.82	95.66	95.85
SCR	VGG16	95.65	91.92	94.36	95.40	95.77
SCR-SRMP	VGG16	95.65	92.84	94.80	95.46	95.78
QCFS [ 41 ]	ResNet-18	96.48	92.86	94.92	96.21	96.55

TABLE IV: Comparison of test results among SRMP, SCR-SRMP, SCR and QCFS on CIFAR-10

Fig. 8: Comparison with fixed weight and dynamic

\lambda_{p}

for

\mathcal{L}_{\mathrm{RMPD}}

Panel of Figure 13 — Fig. 9: Distribution of $\sigma/\theta$ in SNN layers in VGG16, ResNet18 and ResNet-20. Frequency of $\sigma/\theta<0.125$ amounts to $84.67\%$ in VGG-16, which indicates that at least $84.67\%$ of neurons satisfy the conditions of Theorem 1 . Similar observations can be obtained for other models.

Model	Base (M)	with CR (M)	$\Delta$ C (M)	$\Delta$ (%)	$\Delta$ P. (K)
VGG-16	666.50	671.36	+4.86	+0.73	+24.2
ResNet-18	1115.78	1120.50	+4.72	+0.42	+4.0
ResNet-20	83.24	84.12	+0.88	+1.06	+0.4
ResNet-34	2327.02	2334.10	+7.08	+0.30	+6.9

TABLE V: Computational cost comparison.

Spiking Neural Networks

Spiking neural networks process information through discrete, time-dependent neural events, aiming for efficient computation and biologically inspired learning.

See all 351 papers

SpikeQ: Spike-Guided Post-Training Quantization for Spiking Neural Networks
Oct 16, 2026Kai Sun, Peibo Duan, Zhanglu Yan +3Spiking Neural Networks Post-Training Quantization
Monash University · National University of Singapore · Northeastern University
Activation-Flexible ANN-to-SNN Conversion with Finite-State Markov Neurons
Sep 24, 2026Ruiyu Jia, Zhuo-Cheng XiaoNeural Network Activation Functions Spiking Neural Networks
New York University Shanghai, Shanghai, China
Sigma-Delta Neural Network Conversion on Loihi 2
May 9, 2025Matthew Brehove, Sadia Anjum Tumpa, Espoir Kyubwa +2Efficient Neural Network Inference Spiking Neural Networks
ChromoLogic LLC Monrovia, California, USA · Pennsylvania State University University Park, Pennsylvania, USA

Oct 16, 2026J/K move · Enter open · S save

SpikeQ: Spike-Guided Post-Training Quantization for Spiking Neural Networks

Kai Sun, Peibo Duan, Zhanglu Yan, Guowei Zhang+2

Monash University · National University of Singapore · Northeastern University

Spiking Neural Networks Post-Training Quantization Neuromorphic Computing Activation Quantization

Open paper AskPDF arXiv

On neuromorphic chips, spiking neural networks (SNNs) consume far less energy than artificial neural networks (ANNs), but the limited memory of these chips requires low-bit synaptic weights and membrane potentials. Existing post-training quantization (PTQ) methods provide them without retraining and, like those for ANNs, minimize numerical errors such as the mean squared error (MSE). In an SNN, however, a quantization error reaches later layers only through the spikes it flips. We find that errors of equal MSE can cause very different accuracy losses, whereas errors that flip the same fraction of spikes cause the same loss. We therefore propose SpikeQ, a spike-guided PTQ framework. For weights, we select the level of each weight by simulating the downstream neurons and minimizing their spike errors. For membrane potentials, most existing methods quantize the state after the reset, whereas neuromorphic chips compare the threshold with the membrane potential stored at a fixed bit width. We target this pre-fire quantization and build a reset-aware grid that places the threshold on a rounding boundary, so rounding never flips the spike at the current time step. We evaluate SpikeQ on seven benchmarks of image classification, semantic segmentation, and object detection. With 2-bit pre-fire membrane potentials, every existing grid collapses, whereas SpikeQ retains most of the full-precision accuracy. With both at 4 bits, the models stay within 2.59 points of full precision, shrink their memory footprint 7.8--7.9 times, and use 6.4--6.9 times less energy.

Abstract

Spiking Neural Networks

Explore similar work

SpikeQ: Spike-Guided Post-Training Quantization for Spiking Neural Networks

Activation-Flexible ANN-to-SNN Conversion with Finite-State Markov Neurons

Sigma-Delta Neural Network Conversion on Loihi 2