Computer chips capable of mimicking the human mind have been unveiled in a new study conducted by an international team of researchers.
In the report, the team, led by researchers from the University of Zurich and ETH Zurich, demonstrate how complex cognitive abilities can be incorporated into electronic systems assembled using the so-called "neuromorphic" chips.
While most approaches in neuroinformatics, or the combination of research of neuroscience and informatics research, are limited to the development of neural network models on conventional computers or aim to simulate complex nerve networks on supercomputers, the new study focuses on developing electronic circuits that are comparable to a real brain in terms of size, speed and energy consumption.
"Our goal is to emulate the properties of biological neurons and synapses directly on microchips," Giacomo Indiveri, a professor at the Institute of Neuroinformatics (INI), of the University of Zurich and ETH Zurich, said in a press release.
One of the major challenges facing the team in their quest to do so was the configuration of networks made of artificial, i.e. neuromorphic, neurons in such a way that they could perform particular tasks -- a feat, the researchers report, they have now succeeded in doing, as seen the neuromorphic system's ability to carry out complex sensorimotor undertakings in real time.
Among the abilities the system exhibits in performing these tasks is both short-term memory and context-dependent decision-making, both of which are typical traits necessary for cognitive tests.
As part of the process, the INI team combined neuromorphic neurons into networks that implemented neural processing modules equivalent to the "finite-state machines" -- a mathematical concept to describe logical processes or computer programs. Because behavior can be formulated as a "finite-state machine," it can be transferred to the neuromorphic hardware in an automated manner, according to the scientists.
"The network connectivity patterns closely resemble structures that are also found in mammalian brains," Indiveri said.
In doing so, the scientists demonstrate for the first time how a real-time hardware neural-processing system where the user dictates the behavior can be constructed.
"Thanks to our method, neuromorphic chips can be configured for a large class of behavior modes," Indiveri concluded, adding that the group's findings are "pivotal for the development of new brain-inspired technologies."
One such application, the researchers hypothsize, might be the combination of the chips with sensory neuromorphic components, such as an artificial cochlea or retina, to create complex cognitive systems that interact with their surroundings in real time.