Connectivity, memory and neuromorphic function in non-biological networks
John J. Boland
School of Chemistry and Centre for Research on Adaptive Nanostructures and Nanodevices (CRANN), Trinity College Dublin, Ireland

July 1, 2015, 4 p.m.


The seminar will discuss scaling and evolution of connectivity in inorganic nanowire networks when exposed to different stimuli. Network junctions control the overall properties of any network. We discuss how these junctions evolve from capacitor-like to resistor-like barriers in response to a stimulus and how the network self-selects connectivity paths by choosing the lowest barrier junctions. Continued stressing causes the selected junctions to strengthen and ultimately co-opts neighbouring junctions to reinforce the overall response to the stimulus. By engineering these junctions it is possible to control the properties and response of the network. Materials with arbitrarily controlled connectivity and conductivity are demonstrated as are device with arbitrarily controlled multi-level memory, and single junctions capable of a learning response.



Share
Connectivity, memory and neuromorphic function in non-biological networks
John J. Boland
School of Chemistry and Centre for Research on Adaptive Nanostructures and Nanodevices (CRANN), Trinity College Dublin, Ireland

July 1, 2015, 4 p.m.


The seminar will discuss scaling and evolution of connectivity in inorganic nanowire networks when exposed to different stimuli. Network junctions control the overall properties of any network. We discuss how these junctions evolve from capacitor-like to resistor-like barriers in response to a stimulus and how the network self-selects connectivity paths by choosing the lowest barrier junctions. Continued stressing causes the selected junctions to strengthen and ultimately co-opts neighbouring junctions to reinforce the overall response to the stimulus. By engineering these junctions it is possible to control the properties and response of the network. Materials with arbitrarily controlled connectivity and conductivity are demonstrated as are device with arbitrarily controlled multi-level memory, and single junctions capable of a learning response.



Share