• Aucun résultat trouvé

Dynamic Image Quantization using Leaky Integrate-and-Fire Neurons

N/A
N/A
Protected

Academic year: 2021

Partager "Dynamic Image Quantization using Leaky Integrate-and-Fire Neurons"

Copied!
11
0
0

Texte intégral

Loading

Figure

Fig. 1: General framework of the proposed architecture. A 3 × 3 input image I is fed to a group of 9 neurons (in blue).
Fig. 2: LIF model with observation window T and threshold θ. If the intensity I satisfies RI > θ, the neuron spikes (case I ∈ {I 2 , I 3 }), otherwise it remains silent (case I = I 1 ).
Fig. 3: The input values I is arranged in quantization regions S k depending on the number k of emitted spikes.
Fig. 4: (a) Impact of the threshold θ on the performance of the Dual-SIMQ for a zero-mean normal distribution with σ = 2;
+5

Références

Documents relatifs

This work generalizes the use of duality in the theory of Markov processes as proposed by Sigmund (see Anderson’s book). The dual doesn’t

We consider now the left Hilbert algebra associated with the left regular repre- sentation of a locally compact group G endowed with a left Haar measure λ.. The canonical weight τ

Bounded cohomology, classifying spaces, flat bundles, Lie groups, subgroup distortion, stable commutator length, word metrics.. The authors are grateful to the ETH Z¨ urich, the MSRI

LEMMA 1.. - The definitions of g^t) and g\t) and all the statements of Lemma 3 are valid for an arbitrary measurable function g, on an arbitrary measure space (see [3]), but

In this paper, we suggested dynamic programming model for dual crane stocker scheduling problem and developed the model into the deep Q network (DQN).. We obtained an optimal

Our results provide precise orders in the number of mea- surements for the accuracy of the solution to the convex dual problem and an ℓ ∞ error bound on the primal spike

• In Section 3 we define the anti-Hurwitz polynomials, demonstrate our first main result, and establish explicit criteria for real polynomials of less than or equal order 4

Il existe une réelle progression des apprentissages entre le temps 1 (test initial) et le temps 2 (post-test) et ce pour les deux groupes d’élèves, mais cette différence