D. Meunier and H. Paugam-Moisy (2005)
"Evolutionary supervision of a dynamical neural network allows learning with on-going weights."
Proceedings of International Joint Conference on Neural Networks (IJCNN) pp. 1493-1498
==================================================================================================

"Such a modelling of biological data implies that the weight of the synapse is updated at each time the neuron (either pre- or post-synaptic) emits a spike [39][26].

--------------------------------------------------------------------------------------------------
[39] T. Nowotny, V. P. Zigulin, A. I. Selverston, H. D. I. Abarbannel, and M. I. Rabinovitch. "Enhancement of synchronization in a hybrid neural circuit by spiketiming dependant plasticity." J. Neurosci., 23(30):9776.9785, 2003. => got

[26] Eugene M. Izhikevich, Joseph A. Gally, and Gerald M. Edelman. "Spike-timing dynamics of neuronal groups." Cerebral Cortex, 14(8):933.944, 2004. => got
--------------------------------------------------------------------------------------------------

"The classical version of STDP [15, 1, 16] is based on delay selection, and involves a discontinuity between potentiation and depression zones of temporal window : the synapse emitting a PSP that will be responsible for the spike emission is the most potentiated, whereas a PSP arriving just after a spike will be the most deprecated.

--------------------------------------------------------------------------------------------------
[15] W. Gerstner, R. Kempter, J.L. van Hemmen, and H.Wagner. A neuronal learning
rule for sub-millisecond temporal coding. Nature, 383:76.78, 1996.
[16] W. Gerstner and W. Kistler. Spiking Neuron Models: Single Neurons, Populations,
Plasticity. Cambridge University Press, 2002.
[1] L. F. Abbott and S. B. Nelson. Synaptic plastictiy - Taming the beast. Nature
Neurosci., 3:1178.1183, 2000.
--------------------------------------------------------------------------------------------------

Up to now, nobody (to the best of our knowledge) has been able to show how it is possible to learn with STDP, although there are growing evidence that such a mechanism is quite general in the nervous system [9].