By looking at nature, one finds that the mammalian brain is one of (if not) the most efficient network of processing elements (currently) known to mankind: on the order of 1010 processing elements (i.e., neurons) and about 1014 connections (i.e., synapses) — let alone that each synapse has an assoc…
By looking at nature, one finds that the mammalian brain is one of (if not) the most efficient network of processing elements (currently) known to mankind: on the order of 1010 processing elements (i.e., neurons) and about 1014 connections (i.e., synapses) — let alone that each synapse has an associated (low-precision analog) weight. There is no doubt that mammals’ brains have evolved to operate efficiently — the unknown(s) being the optimization cost(s). Obviously, a smaller brain would require fewer materials and less energy for construction, maintenance, and operation.
Just like the wires connecting components in semiconductor chips, the connections between neurons occupy a substantial fraction of the total volume, and the "wires" (i.e., axons and dendrites) are expensive to operate as dissipating energy during signaling. In fact, although the human brain represents only 2% of the total body weight, it consumes 20% of its resting energy — which is obviously quite expensive! However, nature has an important advantage over electronic circuits. In mammals’ brains components are connected in 3D space, whereas even the most advanced chips use only a small number of layers of planar wiring. This could be (one of) the reason(s) why wiring occupies 40% to 60% of the whole brain’s volume — which is still significantly less than the 90% of today’s VLSI chips.
Currently, the fields of computing and neuroscience are interacting in ways nobody could imagine a few years ago. That is why, this workshop aims to bring together researchers with expertise in nano-electronics, as well as neuro-science, but also with particular emphasis on neural communications. These should allow for a cross-disciplinary approach and we do hope to attract researchers who normally go to quite different conferences.