MATLAB : Fourier transform
In the field of AI, researchers are working on devices
having their own memory and ability of thinking, they are trying to mimic the human
brain. The human brain is the most complicated structure consisting of millions
of neurons that are interlinked by a thread-like structure called synapses.
The engineers from the Massachusetts Institute of Technology
have designed 'brain-on -a-chip'. The chip is smaller than a piece of confetti,
that is about 1 mm in dimension and made from thousands of artificial brain
synapses known as Memristors. Memristors are the silicon-based component that
mimics the information-transmitting synapses in the human brain.
Memristors or memory resistors are the necessary elements of
neuromorphic computing. In a neuromorphic device, memristors would work more
similar to a brain synapse. The synapse receives signals from one neuron, in
the form of ions, and sends a corresponding signal to the next neuron. Like a
brain synapse, a memristor would also be able to remember the value of
associated with given current strength and produce exactly the same signal
the next time it receives a similar current.
However, the existing memristor designs are limited in their
performance. A single memristor is made of a positive and negative electrode,
separated by a switching medium, or space between the electrodes. when a
voltage applied to one electrode, ions from that electrode flow through the
medium and forming the conduction channel to the other electrode. The received
ions produce the electrical signals that the memristor transmits through the
circuit.
The team of the researcher first fabricated a negative
electrode out of silicon, then made a positive electrode by depositing a slight
amount of copper, followed by a layer of silver. They sandwiched the two
electrodes around an amorphous silicon medium. In this way, they patterned a
millimetre-square silicon chip with tens of thousands of memristor.
Kim a researcher from MIT says that existing memristor design works pretty well in the case where voltage stimulates a large conduction channel or a heavy flow of ions from one electrode to other. But these designs are less reliable when memristors need to generate narrow signals, via thinner conduction channel.
Memristor could advance the development of small, portable AI devices.
-by Anshul Shakya
original written by Jennifer Chu.
source:- Massachusetts Insitute of Technology
Great!!
ReplyDelete