We examine the ability of combining symbolic and subsymbolic approaches b y means of recursively encoding and decoding structured data. We show that encoding of symbolic data is possible in this w ay { hence neural netw orks seem well suited for control or classi cation in symbolic approaches { whereas decoding requires an increasing complexit y of the decoding function { hence netw orks with t...

We analyze complexity of sleep-EEG-signals using methods based on Higuchi’s fractal dimension and of the signal and a new symbolic method applied to signal’s derivative. We demonstrate that these methods may be helpful in diagnosis of sleep disorders the fractal dimension method differentiate between sleep stages while the symbolic method may differentiate between healthy subject and insomniacs...

Symbolic synchronization and the detection of global properties of coupled dynamics from local information.

We find that the global symbolic dynamics of a diffusively coupled map lattice is well approximated by a very small local model for weak to moderate coupling strengths. A local symbolic model is a truncation of the full symbolic model to one that considers only a single element and a few neighbors. Using interval analysis, we give rigorous results for a range of coupling strengths and different...

Notes for the summer school Symbolic dynamics and homeomorphisms of the Cantor set,

The dynamics of human electroencephalography (EEG) have been proved to be related to cognitive activities. This study separately assessed the two EEG components, amplitude and rhythm, aiming to capture their individual contributions to cognitive functions. We extracted the local peaks of EEGs under rest or photic stimulation and calculated the symbolic dynamics of their voltages (amplitude) and...