Symbiosis Brain - Computer: The First Attempts To Fuse A Neuron With A Microcircuit - Alternative View

Symbiosis Brain - Computer: The First Attempts To Fuse A Neuron With A Microcircuit - Alternative View
Symbiosis Brain - Computer: The First Attempts To Fuse A Neuron With A Microcircuit - Alternative View

Video: Symbiosis Brain - Computer: The First Attempts To Fuse A Neuron With A Microcircuit - Alternative View

Video: Symbiosis Brain - Computer: The First Attempts To Fuse A Neuron With A Microcircuit - Alternative View
Video: Max Bennett: Decoding How Intelligence Is Implemented in the Brain 2024, November
Anonim

Scientists from the National Research Nizhny Novgorod State University named after NI Lobachevsky (participant of Project 5-100) are working on the creation of an adaptive neurointerface consisting of conjugated neural networks of the brain and electronic neuromorphic systems based on memristors.

This work is one of the first attempts to combine living biological culture with a biosimilar neural network based on memristors.

In 1971, a memristor was first mentioned in an article by University of California professor Leon Chua. Chua theoretically predicted the presence of another element of electrical circuits along with resistance, inductance and capacitance, calling it "memristor".

A memristive chip in a package placed in a standard contacting device (for testing the parameters of memristive nanostructures) / E. Emelyanova (NIPT NNSU)
A memristive chip in a package placed in a standard contacting device (for testing the parameters of memristive nanostructures) / E. Emelyanova (NIPT NNSU)

A memristive chip in a package placed in a standard contacting device (for testing the parameters of memristive nanostructures) / E. Emelyanova (NIPT NNSU)

This definition is still debated, as they argue about the fundamental question: what is a memristor? Initially it was invented as a new element of electrical circuits, but now many believe that the memristor carries a certain expansion of the functionality of the resistor: the memristor is a "resistor with memory."

Unlike a conventional resistor (resistance), which determines the linear dependence of current on voltage, a memristor is a nonlinear element, the resistance of which depends on the "prehistory", for example, on what current flowed through it. He kind of "remembers" what was passed through him, and his state changes depending on this. This adaptive behavior of the memristor is very similar to what we can find in nature, in particular, in the nervous system, where the synapse plays this role. Accordingly, biosimilar memristor systems are systems for which a memristor is the basic element.

As for the devices of such systems, there may be different approaches, and the scientists of the UNN offer their own version.

The scheme of combining living neurons with a neural network based on memristors (UNN) / UNN
The scheme of combining living neurons with a neural network based on memristors (UNN) / UNN

The scheme of combining living neurons with a neural network based on memristors (UNN) / UNN

Promotional video:

An adaptive neurointerface is being developed on the basis of the NIFTI NNSU and the Nizhny Novgorod Neuroscience Center, in which, on the one hand, there is a living culture, and on the other, a neural network based on memristors. Memristor neural networks will be interfaced with a multi-electrode system for recording and stimulating the bioelectric activity of a neuron culture, which performs the function of analyzing and classifying the network dynamics of living cells.

NNSU scientists and equipment for the formation of memristive nanostructures (plasma jet etching unit) / E. Emelyanova (NIPT NNSU)
NNSU scientists and equipment for the formation of memristive nanostructures (plasma jet etching unit) / E. Emelyanova (NIPT NNSU)

NNSU scientists and equipment for the formation of memristive nanostructures (plasma jet etching unit) / E. Emelyanova (NIPT NNSU)

At the moment, scientists are exploring the possibility of constructing a feedback, within which the output signal from the memristor network will be used to stimulate the biological network - that is, the learning process of a living cell culture is being implemented for the first time.

Prototype of an artificial neural network based on a hybrid analog-digital electronic circuit and a memristive chip / E. Emelyanova (NIPT NNSU)
Prototype of an artificial neural network based on a hybrid analog-digital electronic circuit and a memristive chip / E. Emelyanova (NIPT NNSU)

Prototype of an artificial neural network based on a hybrid analog-digital electronic circuit and a memristive chip / E. Emelyanova (NIPT NNSU)

As a living culture, scientists use artificially grown neuronal brain cell culture. But, in principle, a slice of living tissue can also be used.

Compared to global competitors who set the task of "connecting the living world and artificial architectures" (for example, the RAMP project), the advantage of the UNN project is that qualified specialists in various fields - physics and technology for creating memristive nanostructures, modeling neural networks, designing electronic schemes, neurodynamics and neurobiology - are concentrated both geographically and organizationally within one university.

Head of the Laboratory of Physics and Technology of Thin Films, NIPhT NNSU named after N. I. Lobachevsky, Candidate of Physical and Mathematical Sciences Alexey Mikhailov / E. Emelyanova (NIPT NNSU)
Head of the Laboratory of Physics and Technology of Thin Films, NIPhT NNSU named after N. I. Lobachevsky, Candidate of Physical and Mathematical Sciences Alexey Mikhailov / E. Emelyanova (NIPT NNSU)

Head of the Laboratory of Physics and Technology of Thin Films, NIPhT NNSU named after N. I. Lobachevsky, Candidate of Physical and Mathematical Sciences Alexey Mikhailov / E. Emelyanova (NIPT NNSU)

The head of the Laboratory of Physics and Technology of Thin Films of the NIPTI NNSU named after NI Lobachevsky, Candidate of Physical and Mathematical Sciences Alexei Mikhailov explains: “We are trying to create a prototype of a neural network based on memristors, which in its internal structure and functionality is similar to the biological nervous system. Due to the locality of the memristive effect (the corresponding phenomena occur on a nanoscale) and the use of modern standard microelectronic technologies, it will be possible to obtain a large number of neurons and synapses on one chip. This is a distant prospect that we are striving for. That is, a human brain can be “grown” on a crystal, on a chip. While we are making things simpler: we are trying to create hybrid electronic circuits in which some functions are implemented on the basis of traditional electronics (transistors),and some new functions that are difficult to implement in hardware are provided on the basis of memristors."

UNN scientists in the course of studying the parameters of the adaptive response of memristive devices / E. Emelyanova (NIPT NNSU)
UNN scientists in the course of studying the parameters of the adaptive response of memristive devices / E. Emelyanova (NIPT NNSU)

UNN scientists in the course of studying the parameters of the adaptive response of memristive devices / E. Emelyanova (NIPT NNSU)

The goal of the project is to create compact electronic devices based on memristors that reproduce the property of synaptic plasticity and function as part of biosimilar neural networks in conjunction with living biological cultures. The use of hybrid neural networks based on memristors opens up amazing prospects. First, a memristor can help fit the power of modern supercomputers on a single chip. Secondly, it will be possible to create robots that control artificially grown neural cultures. Third, such "brain-like" systems can be used to replace a part of the living nervous system with an electronic one in case of damage or disease.