Neural Computing with Concurrent Synchrony

Neural networks are important modeling tools to implement intelligent behaviour in a wide variety of phenomena. We introduce the concept of concurrent synchrony in spikes to enable the efficient representation of neural networks to process sensory stimuli. Using different sensory modalities, we show that information processing from stimuli can be represented compactly. This approach aims at introducing homeostasis into the behavior of neural populations in order to construct diverse and sophisticated control rules without increasing network complexity.