Scientists have found a material that gives AI the ability to mimic the human brain.
In the light of the rapid development of AI technologies, there is a need for a new generation of hardware that can cope with increasing loads. Modern computer architectures based on silicon are not adapted to support such technologies.
In response to this challenge, a team of scientists is investigating neuromorphic architectures that mimic the functions of the human brain and can handle AI calculations more efficiently due to lower power consumption and improved computing capabilities.
One of the materials suitable for creating neuromorphic devices is vanadium dioxide VO2, which can change its electrical properties — be a dielectric (insulator)at one temperature and a metal at another. This property makes vanadium dioxide an ideal candidate for creating artificial neurons and synapses, which is key for simulating the human brain in computing.
Interestingly, vanadium dioxide memory is distributed over the entire volume of the material, which significantly increases the storage density of data and provides new ways to manage this information. The research opens the door to creating more advanced artificial networks that can mimic complex processes occurring in the brain.
Now, the researchers are planning further experiments that will allow them to fine-tune the local properties of the material and improve its synaptic behavior, which could significantly advance the development of neuromorphic computing. This also includes ion bombardment, which can pinpoint material properties to create more efficient neuromorphic devices.
In the light of the rapid development of AI technologies, there is a need for a new generation of hardware that can cope with increasing loads. Modern computer architectures based on silicon are not adapted to support such technologies.
In response to this challenge, a team of scientists is investigating neuromorphic architectures that mimic the functions of the human brain and can handle AI calculations more efficiently due to lower power consumption and improved computing capabilities.
One of the materials suitable for creating neuromorphic devices is vanadium dioxide VO2, which can change its electrical properties — be a dielectric (insulator)at one temperature and a metal at another. This property makes vanadium dioxide an ideal candidate for creating artificial neurons and synapses, which is key for simulating the human brain in computing.
Interestingly, vanadium dioxide memory is distributed over the entire volume of the material, which significantly increases the storage density of data and provides new ways to manage this information. The research opens the door to creating more advanced artificial networks that can mimic complex processes occurring in the brain.
Now, the researchers are planning further experiments that will allow them to fine-tune the local properties of the material and improve its synaptic behavior, which could significantly advance the development of neuromorphic computing. This also includes ion bombardment, which can pinpoint material properties to create more efficient neuromorphic devices.
