Reservoir computing systems built
with memristors, however, can skip most of the expensive training process and still provide the network the capability to remember.
Tuszynski wasn't familiar
with memristors until he met Chua at a 2015 conference in India.
Not exact matches
With combined characteristics of a
memristor and transistor, the memtransistor also encompasses multiple terminals that operate more similarly to a neural network.
«Memtransistor» brings world closer to brain - like computing: Combined
memristor and transistor can process information and store memory
with one device.»
The
memristor fills this hole, creating a fourth circuit element that would operate like a resistor
with memory.
To test the
memristor theory, Tuszynski's team filled a dish
with microtubules, tubulin proteins and a buffer solution, then added electricity.
For one, «HP's
memristor is made using a titanium oxide, which is not common to put on semiconductors and appears to be difficult to manage today,» says Jim Handy, an analyst
with the semiconductor market research firm Objective Analysis in Los Gatos, Calif..
A
memristor - based machine could perform some tasks, like voice or face recognition, much better than any machine today, Williams says, adding, «At least in theory, a machine [based on
memristors] could do those things almost instantaneously
with no waste of time or energy.»
For decades, the
memristor — short for memory resistor — had been little more than an engineer's dream: a circuit element
with high - density memory and low energy requirements.
The Prodromakis Group at the University of Southampton is acknowledged as world - leading in this field, collaborating among others
with Leon Chua (a Diamond Jubilee Visiting Academic at the University of Southampton), who theoretically predicted the existence of
memristors in 1971.
But it works well as it is, so we are also thinking of building a more complex system
with such spin -
memristors to test actual algorithms for specific cognition capabilities of the human brain.»
The device developed by the physicists combines the
memristor effect of semiconductors
with a spin - based phenomenon called tunnelling anisotropic magnetoresistance (TAMR) and works at room temperature.
The artificial network is based on an advanced chip called a «
memristor» (an electrical resistor
with memory), which stores and processes data simultaneously and enables efficient pattern recognition.
The work is featured in the advance online publication of Nature Materials («
Memristors with diffusive dynamics as synaptic emulators for neuromorphic computing»).
Home > Press > «Memtransistor» brings world closer to brain - like computing: Combined
memristor and transistor can process information and store memory
with one device
«Memtransistor» brings world closer to brain - like computing: Combined
memristor and transistor can process information and store memory
with one device February 22nd, 2018