Discrete-time LIF spiking networks realize piecewise constant functions on polyhedral regions, and each first-layer neuron generates only O(T^2) parallel hyperplanes over T time steps, not exponentially many.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Time to Spike? Understanding the Representational Power of Spiking Neural Networks in Discrete Time
Discrete-time LIF spiking networks realize piecewise constant functions on polyhedral regions, and each first-layer neuron generates only O(T^2) parallel hyperplanes over T time steps, not exponentially many.