The NCS group was established in early 2009 at the Institute of Neuroinformatics to study and develop computational models, hybrid analog/digital VLSI circuits, and multi-chip event-based systems for implementing Neuromorphic Cognitive Systems.
What is a neuromorphic cognitive system and what is neuromorphic cognition?
We don't have a comprehensive answer to these questions yet. But our work is aimed at bridging the gap from simple reactive neuromorphic sensory-motor systems to ones that are cognitive in quality: e.g. neuromorphic architectures that can reason about the actions to take, in response to the combinations of external stimuli, internal states, and behavioral objectives.
We focus on analog/digital VLSI architectures that use the physics of silicon to reproduce the biophysics of biological neural systems, and multi-chip systems that communicate using asynchronous event-based signals (spikes).
Our main contributions to neuromorphic VLSI technology consist of novel low-power silicon neuron circuits, dynamic silicon synapses, hybrid analog/digital spike-based learning mechanisms, soft Winner-take-all networks, and asynchronous digital communication circuits and systems.