Neural Networks
Spiking Neuron Models: An Introduction
Spiking Neuron Models: An Introduction
A population study of integrate-and-fire-or-burst neurons
Neural Computation
Isolated word recognition using high-order statistics and time-delay neural networks
SPWHOS '97 Proceedings of the 1997 IEEE Signal Processing Workshop on Higher-Order Statistics (SPW-HOS '97)
A complex-valued spiking machine
ICANN/ICONIP'03 Proceedings of the 2003 joint international conference on Artificial neural networks and neural information processing
| Hi-index | 0.00 |
The temporal dimension is very important to be considered in many cognitive tasks involving a decision making or a behavior in response to spatio-temporal stimuli, such as vision, speech and signal processing. Thus, the capacity of encoding, recognizing, and recalling spatio-temporal patterns is one of the most crucial features of any intelligent system either artificial or biologic. If some connexionnist or hybrid model integrates the temporal data as spatial input, few other models take them into account together internally either in training or in architecture. Temporal Organization Map TOM is one of the latest types. In this paper, we propose a model gathering saptio-temporal data coding, representation and processing based on TOM map, and yielding to a Spatio-Temporel Organization Map (STOM). For spatio-temporal data coding, we use the domain of complex numbers to represent the two dimensions together. STOM architecture is the same as TOM, however, training is ensured by the spatio-temporal Kohonen algorithm to make it able to manage complex input.