Tag Archives: biological neurons

Nano-neurons from a French-Japanese-US research team

This news about nano-neurons comes from a Nov. 8, 2017 news item on defenceweb.co.za,

Researchers from the Joint Physics Unit CNRS/Thales, the Nanosciences and Nanotechnologies Centre (CNRS/Université Paris Sud), in collaboration with American and Japanese researchers, have developed the world’s first artificial nano-neuron with the ability to recognise numbers spoken by different individuals. Just like the recent development of electronic synapses described in a Nature article, this electronic nano-neuron is a breakthrough in artificial intelligence and its potential applications.

A Sept. 19, 2017 Thales press release, which originated the news item, expands on the theme,

The latest artificial intelligence algorithms are able to recognise visual and vocal cues with high levels of performance. But running these programs on conventional computers uses 10,000 times more energy than the human brain. To reduce electricity consumption, a new type of computer is needed. It is inspired by the human brain and comprises vast numbers of miniaturised neurons and synapses. Until now, however, it had not been possible to produce a stable enough artificial nano-neuron which would process the information reliably.

Today [Sept. 19, 2017 or July 27, 2017 when the paper was published in Nature?]], for the first time, researchers have developed a nano-neuron with the ability to recognise numbers spoken by different individuals with 99.6% accuracy. This breakthrough relied on the use of an exceptionally stable magnetic oscillator. Each gyration of this nano-compass generates an electrical output, which effectively imitates the electrical impulses produced by biological neurons. In the next few years, these magnetic nano-neurons could be interconnected via artificial synapses, such as those recently developed, for real-time big data analytics and classification.

The project is a collaborative initiative between fundamental research laboratories and applied research partners. The long-term goal is to produce extremely energy-efficient miniaturised chips with the intelligence needed to learn from and adapt to the constantly ever-changing and ambiguous situations of the real world. These electronic chips will have many practical applications, such as providing smart guidance to robots or autonomous vehicles, helping doctors in their diagnosis’ and improving medical prostheses. This project included researchers from the Joint Physics Unit CNRS/Thales, the AIST, the CNS-NIST, and the Nanosciences and Nanotechnologies Centre (CNRS/Université Paris-Sud).

About the CNRS
The French National Centre for Scientific Research is Europe’s largest public research institution. It produces knowledge for the benefit of society. With nearly 32,000 employees, a budget exceeding 3.2 billion euros in 2016, and offices throughout France, the CNRS is present in all scientific fields through its 1100 laboratories. With 21 Nobel laureates and 12 Fields Medal winners, the organization has a long tradition of excellence. It carries out research in mathematics, physics, information sciences and technologies, nuclear and particle physics, Earth sciences and astronomy, chemistry, biological sciences, the humanities and social sciences, engineering and the environment.

About the Université Paris-Saclay (France)
To meet global demand for higher education, research and innovation, 19 of France’s most renowned establishments have joined together to form the Université Paris-Saclay. The new university provides world-class teaching and research opportunities, from undergraduate courses to graduate schools and doctoral programmes, across most disciplines including life and natural sciences as well as social sciences. With 9,000 masters students, 5,500 doctoral candidates, an equivalent number of engineering students and an extensive undergraduate population, some 65,000 people now study at member establishments.

About the Center for Nanoscale Science & Technology (Maryland, USA)
The CNST is a national user facility purposely designed to accelerate innovation in nanotechnology-based commerce. Its mission is to operate a national, shared resource for nanoscale fabrication and measurement and develop innovative nanoscale measurement and fabrication capabilities to support researchers from industry, academia, NIST and other government agencies in advancing nanoscale technology from discovery to production. The Center, located in the Advanced Measurement Laboratory Complex on NIST’s Gaithersburg, MD campus, disseminates new nanoscale measurement methods by incorporating them into facility operations, collaborating and partnering with others and providing international leadership in nanotechnology.

About the National Institute of Advanced Industrial Science and Technology (Japan)
The National Institute of Advanced Industrial Science and Technology (AIST), one of the largest public research institutes in Japan, focuses on the creation and practical realization of technologies useful to Japanese industry and society, and on bridging the gap between innovative technological seeds and commercialization. For this, AIST is organized into 7 domains (Energy and Environment, Life Science and Biotechnology, Information Technology and Human Factors, Materials and Chemistry, Electronics and Manufacturing, Geological

About the Centre for Nanoscience and Nanotechnology (France)
Established on 1 June 2016, the Centre for Nanosciences and Nanotechnologies (C2N) was launched in the wake of the joint CNRS and Université Paris-Sud decision to merge and gather on the same campus site the Laboratory for Photonics and Nanostructures (LPN) and the Institut d’Electronique Fondamentale (IEF). Its location in the École Polytechnique district of the Paris-Saclay campus will be completed in 2017 while the new C2N buildings are under construction. The centre conducts research in material science, nanophotonics, nanoelectronics, nanobiotechnologies and microsystems, as well as in nanotechnologies.

There is a video featuring researcher Julie Grollier discussing their work but you will need your French language skills,

(If you’re interested, there is an English language video published on youtube on Feb. 19, 2017 with Julie Grollier speaking more generally about the field at the World Economic Forum about neuromorphic computing,  https://www.youtube.com/watch?v=Sm2BGkTYFeQ

Here’s a link to and a citation for the team’s July 2017 paper,

Neuromorphic computing with nanoscale spintronic oscillators by Jacob Torrejon, Mathieu Riou, Flavio Abreu Araujo, Sumito Tsunegi, Guru Khalsa, Damien Querlioz, Paolo Bortolotti, Vincent Cros, Kay Yakushiji, Akio Fukushima, Hitoshi Kubota, Shinji Yuasa, Mark D. Stiles, & Julie Grollier. Nature 547, 428–431 (27 July 2017) doi:10.1038/nature23011 Published online 26 July 2017

This paper is behind a paywall.

Memristive capabilities from IBM (International Business Machines)

Does memristive mean it’s like a memristor but it’s not one? In any event, IBM is claiming some new ground in the world of cognitive computing (also known as, neuromorphic computing).

An artistic rendering of a population of stochastic phase-change neurons which appears on the cover of Nature Nanotechnology, 3 August 2016. (Credit: IBM Research)

An artistic rendering of a population of stochastic phase-change neurons which appears on the cover of Nature Nanotechnology, 3 August 2016. (Credit: IBM Research)

From an Aug. 3, 2016 news item on phys.org,

IBM scientists have created randomly spiking neurons using phase-change materials to store and process data. This demonstration marks a significant step forward in the development of energy-efficient, ultra-dense integrated neuromorphic technologies for applications in cognitive computing.

Inspired by the way the biological brain functions, scientists have theorized for decades that it should be possible to imitate the versatile computational capabilities of large populations of neurons. However, doing so at densities and with a power budget that would be comparable to those seen in biology has been a significant challenge, until now.

“We have been researching phase-change materials for memory applications for over a decade, and our progress in the past 24 months has been remarkable,” said IBM Fellow Evangelos Eleftheriou. “In this period, we have discovered and published new memory techniques, including projected memory, stored 3 bits per cell in phase-change memory for the first time, and now are demonstrating the powerful capabilities of phase-change-based artificial neurons, which can perform various computational primitives such as data-correlation detection and unsupervised learning at high speeds using very little energy.”

An Aug. 3, 2016 IBM news release, which originated the news item, expands on the theme,

The artificial neurons designed by IBM scientists in Zurich consist of phase-change materials, including germanium antimony telluride, which exhibit two stable states, an amorphous one (without a clearly defined structure) and a crystalline one (with structure). These materials are the basis of re-writable Blu-ray discs. However, the artificial neurons do not store digital information; they are analog, just like the synapses and neurons in our biological brain.

In the published demonstration, the team applied a series of electrical pulses to the artificial neurons, which resulted in the progressive crystallization of the phase-change material, ultimately causing the neuron to fire. In neuroscience, this function is known as the integrate-and-fire property of biological neurons. This is the foundation for event-based computation and, in principle, is similar to how our brain triggers a response when we touch something hot.

Exploiting this integrate-and-fire property, even a single neuron can be used to detect patterns and discover correlations in real-time streams of event-based data. For example, in the Internet of Things, sensors can collect and analyze volumes of weather data collected at the edge for faster forecasts. The artificial neurons could be used to detect patterns in financial transactions to find discrepancies or use data from social media to discover new cultural trends in real time. Large populations of these high-speed, low-energy nano-scale neurons could also be used in neuromorphic coprocessors with co-located memory and processing units.

IBM scientists have organized hundreds of artificial neurons into populations and used them to represent fast and complex signals. Moreover, the artificial neurons have been shown to sustain billions of switching cycles, which would correspond to multiple years of operation at an update frequency of 100 Hz. The energy required for each neuron update was less than five picojoule and the average power less than 120 microwatts — for comparison, 60 million microwatts power a 60 watt lightbulb.

“Populations of stochastic phase-change neurons, combined with other nanoscale computational elements such as artificial synapses, could be a key enabler for the creation of a new generation of extremely dense neuromorphic computing systems,” said Tomas Tuma, a co-author of the paper.

Here’s a link to and a citation for the paper,

Stochastic phase-change neurons by Tomas Tuma, Angeliki Pantazi, Manuel Le Gallo, Abu Sebastian, & Evangelos Eleftheriou. Nature Nanotechnology  11, 693–699 (2016) doi:10.1038/nnano.2016.70 Published online 16 May 2016

I gather IBM waited for the print version of the paper before publicizing the work. The online version is behind paper. For those who can’t get past the paywall, there is a video offering a demonstration of sorts,

For the interested, the US government recently issued a white paper on neuromorphic computing (my Aug. 22, 2016 post).

This team has published a paper that has a similar theme to the one in Nature Nanotechnology,

All-memristive neuromorphic computing with level-tuned neurons by Angeliki Pantazi, Stanisław Woźniak, Tomas Tuma, and Evangelos Eleftheriou. Nanotechnology, Volume 27, Number 35  DOI: 10.1088/0957-4484/27/35/355205 Published 26 July 2016

© 2016 IOP Publishing Ltd

This paper is open access.

An Aug. 18, 2016 news piece by Lisa Zyga for phys.org provides a summary of the research in the July 2016 published paper.