Upper bounds on the number of causal pieces in spiking neural networks
Abstract
The input space of single-spike neural networks decomposes into causal pieces, and the number of causal pieces has been shown to provide a lower bound on the approximation error for some spiking neural networks. While this decomposition is applicable to a wide range of spiking neuron models, the limits of how many causal pieces can be realised by spiking neural networks are still unknown. In this work, we derive two complementary upper bounds on the number of causal pieces in feedforward single-spike neural networks. First, we establish a combinatorial bound given by the poly-Bernoulli numbers, which applies to a broad class of spiking neuron models. The bound is of order e2(L−1)Nlog N+𝒪(N), with network depth L and maximum layer width N. Second, for a specific set of neuron models, we derive a bound based on hyperplane arrangements. Applying the latter framework, we obtain bounds of order 𝒪 (N(4L−3)(N0−1)) for the non-leaky integrate-and-fire model with exponential synapses and the ReLU-type spike response model, and 𝒪 (NL(N0−1)) for the leaky integrate-and-fire model with Dirac-δ synapses, where N0 is the input dimension. For the non-leaky integrate-and-fire model with Dirac-δ synapses, additional temporal structure allows us to sharpen the general combinatorial bound to one that depends only on N0, independently of the network depth and widths.
Related
- The Spectrum of Asynchronous Dynamics in Spiking Networks as a Model for the Diversity of Non-rhythmic Waking States in the Neocortex
- Fast rate coding in heterogeneous spiking networks
- Stochastic Dynamics of Large-Scale Motif-Embedded Spiking Neuronal Networks
- Evolved Spiking Neural Networks for EEG Seizure Detection: Calcium Dynamics and Delay Mechanics
- Metabolic Cost Asymmetry as a Structural Regulariser for Recurrent Spiking Neural Networks
BibTeX
@misc{dold2026upper,
title = {{Upper bounds on the number of causal pieces in spiking neural networks}},
author = {Dominik Dold and Dennis Elbrächter and Philipp Petersen},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.22484},
eprint = {2610.22484},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.22484},
note = {CC BY 4.0},
}
RIS
TY - UNPB
TI - Upper bounds on the number of causal pieces in spiking neural networks
AU - Dominik Dold
AU - Dennis Elbrächter
AU - Philipp Petersen
PY - 2026
DA - 2026/10/15
PB - arXiv
DO - 10.48550/arxiv.2610.22484
UR - https://arxiv.org/abs/2610.22484
AB - The input space of single-spike neural networks decomposes into causal pieces, and the number of causal pieces has been shown to provide a lower bound on the approximation error for some spiking neural networks. While this decomposition is applicable to a wide range of spiking neuron models, the limits of how many causal pieces can be realised by spiking neural networks are still unknown. In this work, we derive two complementary upper bounds on the number of causal pieces in feedforward single-spike neural networks. First, we establish a combinatorial bound given by the poly-Bernoulli numbers, which applies to a broad class of spiking neuron models. The bound is of order e^2(L−1)Nlog N+𝒪(N), with network depth L and maximum layer width N. Second, for a specific set of neuron models, we derive a bound based on hyperplane arrangements. Applying the latter framework, we obtain bounds of order 𝒪 (N^(4L−3)(N_0−1)) for the non-leaky integrate-and-fire model with exponential synapses and the ReLU-type spike response model, and 𝒪 (N^L(N_0−1)) for the leaky integrate-and-fire model with Dirac-δ synapses, where N_0 is the input dimension. For the non-leaky integrate-and-fire model with Dirac-δ synapses, additional temporal structure allows us to sharpen the general combinatorial bound to one that depends only on N_0, independently of the network depth and widths.
ER -