Synaptic delay learning in pulse-coupled neurons.

We present rules for the unsupervised learning of coincidence between excitatory postsynaptic potentials (EPSPs) by the adjustment of postsynaptic delays between the transmitter binding and the opening of ion channels. Starting from a gradient descent scheme, we develop a robust and more biological threshold rule by which EPSPs from different synapses can be gradually pulled into coincidence. The synaptic delay changes are determined from the summed potential--at the site where the coincidence is to be established--and from postulated synaptic learning functions that accompany the individual EPSPs. According to our scheme, templates for the detection of spatiotemporal patterns of synaptic activation can be learned, which is demonstrated by computer simulation. Finally, we discuss possible relations to biological mechanisms.

Correlative velocity estimation: visual motion analysis, independent of object form, in arrays of velocity-tuned bilocal detectors.

The visual estimation of object velocity in systems of tuned bilocal detector units (simplified Hassenstein-Reichardt detectors) is investigated. The units contain delay filters of an arbitrary low-pass characteristic. Arrays of such detector units with identical delay filters are assumed to cover the plane of analysis. The global evaluation of the output signals of suitably arranged detector units is exemplified by the analysis of frontoparallel translations of rigid objects. The correlative method permits the estimation of the instantaneous object velocity, independently of object form. The time course of the resulting estimate is shown to be the convolution of the true velocity profile with a time-invariant kernel that depends solely on the impulse response of the delay filters and thus characterizes the analyzer system. The mathematical analysis of the processing principle is illustrated by considering idealized detector systems. The response of correlative motion analyzers to compound motion and to motion of nonrigid objects is discussed.

On functional concepts for the explanation of visual pattern recognition.

An overview is given covering approximately 30 years of research in the field of functional concepts for the explanation of visual pattern recognition processes. Due to its important role, the Perceptron approach is briefly sketched. This concept, representing to a great extent the present opinion about pattern recognition in the field of visual sciences, is investigated with respect to its invariance properties and its structure. Fundamental facts about the nature of human perception (Gestalt-aspect) and about the processing structure of the visual system are used as qualification criteria. Alternatively, a recent concept that explains the high degree of invariance in visual perception by the evaluation of inner pattern relations is introduced and investigated with respect to its biological plausibility. It is also compared with the Perceptron concept.

Neural computation of inner geometric pattern relations.

A method for the description of patterns is proposed that is based on the evaluation of their inner geometric relations. They serve as features and are determined through operations that are mathematically formulated by so-called "generalized auto comparison functions", i.e., by measures that express a pattern's "auto-match" under geometric transformations. A subset of these features, namely the similarity features, are treated in greater detail, especially with regard to their invariance properties. The dominant role of spatial relations in the formation process of early visual representations is exemplified and a mechanism for the extraction of relational features from such representations is proposed. The feasibility for self-organization of suitable computing structures is discussed.