Tag Archives: neuromorphic hardware

A hardware (neuromorphic and quantum) proposal for handling increased AI workload

It’s been a while since I’ve featured anything from Purdue University (Indiana, US). From a November 7, 2023 news item on Nanowerk, Note Links have been removed,

Technology is edging closer and closer to the super-speed world of computing with artificial intelligence. But is the world equipped with the proper hardware to be able to handle the workload of new AI technological breakthroughs?
Key Takeaways
Current AI technologies are strained by the limitations of silicon-based computing hardware, necessitating new solutions.
Research led by Erica Carlson [Purdue University] suggests that neuromorphic [brainlike] architectures, which replicate the brain’s neurons and synapses, could revolutionize computing efficiency and power.
Vanadium oxides have been identified as a promising material for creating artificial neurons and synapses, crucial for neuromorphic computing.
Innovative non-volatile memory, observed in vanadium oxides, could be the key to more energy-efficient and capable AI hardware.
Future research will explore how to optimize the synaptic behavior of neuromorphic materials by controlling their memory properties.
…

The colored landscape above shows a transition temperature map of VO2 (pink surface) as measured by optical microscopy. This reveals the unique way that this **neuromorphic quantum material** [emphasis mine] stores memory like a synapse. Image credit: Erica Carlson, Alexandre Zimmers, and Adobe Stock

An October 13, 2023 Purdue University news release (also on EurekAlert but published November 6, 2023) by Cheryl Pierce, which originated the news item, provides more detail about the work, Note: A link has been removed,

“The brain-inspired codes of the AI revolution are largely being run on conventional silicon computer architectures which were not designed for it,” explains Erica Carlson, 150^th Anniversary Professor of Physics and Astronomy at Purdue University.
A joint effort between Physicists from Purdue University, University of California San Diego (USCD) and École Supérieure de Physique et de Chimie Industrielles (ESPCI) in Paris, France, believe they may have discovered a way to rework the hardware…. [sic] By mimicking the synapses of the human brain. They published their findings, “Spatially Distributed Ramp Reversal Memory in VO₂” in Advanced Electronic Materials which is featured on the back cover of the October 2023 edition.
New paradigms in hardware will be necessary to handle the complexity of tomorrow’s computational advances. According to Carlson, lead theoretical scientist of this research, “neuromorphic architectures hold promise for lower energy consumption processors, enhanced computation, fundamentally different computational modes, native learning and enhanced pattern recognition.”
Neuromorphic architecture basically boils down to computer chips mimicking brain behavior. Neurons are cells in the brain that transmit information. Neurons have small gaps at their ends that allow signals to pass from one neuron to the next which are called synapses. In biological brains, these synapses encode memory. This team of scientists concludes that vanadium oxides show tremendous promise for neuromorphic computing because they can be used to make both artificial neurons and synapses.
“The dissonance between hardware and software is the origin of the enormously high energy cost of training, for example, large language models like ChatGPT,” explains Carlson. “By contrast, neuromorphic architectures hold promise for lower energy consumption by mimicking the basic components of a brain: neurons and synapses. Whereas silicon is good at memory storage, the material does not easily lend itself to neuron-like behavior. Ultimately, to provide efficient, feasible neuromorphic hardware solutions requires research into materials with radically different behavior from silicon – ones that can naturally mimic synapses and neurons. Unfortunately, the competing design needs of artificial synapses and neurons mean that most materials that make good synaptors fail as neuristors, and vice versa. Only a handful of materials, most of them quantum materials, have the demonstrated ability to do both.”
The team relied on a recently discovered type of non-volatile memory which is driven by repeated partial temperature cycling through the insulator-to-metal transition. This memory was discovered in vanadium oxides.
Alexandre Zimmers, lead experimental scientist from Sorbonne University and École Supérieure de Physique et de Chimie Industrielles, Paris, explains, “Only a few quantum materials are good candidates for future neuromorphic devices, i.e., mimicking artificial synapses and neurons. For the first time, in one of them, vanadium dioxide, we can see optically what is changing in the material as it operates as an artificial synapse. We find that memory accumulates throughout the entirety of the sample, opening new opportunities on how and where to control this property.”
“The microscopic videos show that, surprisingly, the repeated advance and retreat of metal and insulator domains causes memory to be accumulated throughout the entirety of the sample, rather than only at the boundaries of domains,” explains Carlson. “The memory appears as shifts in the local temperature at which the material transitions from insulator to metal upon heating, or from metal to insulator upon cooling. We propose that these changes in the local transition temperature accumulate due to the preferential diffusion of point defects into the metallic domains that are interwoven through the insulator as the material is cycled partway through the transition.”
Now that the team has established that vanadium oxides are possible candidates for future neuromorphic devices, they plan to move forward in the next phase of their research.
“Now that we have established a way to see inside this neuromorphic material, we can locally tweak and observe the effects of, for example, ion bombardment on the material’s surface,” explains Zimmers. “This could allow us to guide the electrical current through specific regions in the sample where the memory effect is at its maximum. This has the potential to significantly enhance the synaptic behavior of this neuromorphic material.”

There’s a very interesting 16 mins. 52 secs. video embedded in the October 13, 2023 Purdue University news release. In an interview with Dr. Erica Carlson who hosts The Quantum Age website and video interviews on its YouTube Channel, Alexandre Zimmers takes you from an amusing phenomenon observed by 19th century scientists through the 20th century where it becomes of more interest as the nanscale phenonenon can be exploited (sonar, scanning tunneling microscopes, singing birthday cards, etc.) to the 21st century where we are integrating this new information into a quantum* material for neuromorphic hardware.

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

Spatially Distributed Ramp Reversal Memory in VO2 by Sayan Basak, Yuxin Sun, Melissa Alzate Banguero, Pavel Salev, Ivan K. Schuller, Lionel Aigouy, Erica W. Carlson, Alexandre Zimmers. Advanced Electronic Materials Volume 9, Issue 10 October 2023 2300085 DOI: https://doi.org/10.1002/aelm.202300085 First published: 10 July 2023

This paper is open access.

There’s a lot of research into neuromorphic hardware, here’s a sampling of some of my most recent posts on the topic,

October 12, 2023 posting, IBM’s neuromorphic chip, a prototype and more
June 23, 2023 posting, Neuromorphic engineering: an overview
August 26, 2022 posting, Save energy with neuromorphic (brainlike) hardware
August 23, 2022 posting, Neuromorphic hardware could yield computational advantages for more than just artificial intelligence
January 26, 2022 posting, Highly scalable neuromorphic (brainlike) computing hardware

There’s more, just use ‘neuromorphic hardware’ for your search term.

*’meta’ changed to ‘quantum’ on January 8, 2024.

Brainlike computing with spintronic devices

Brainy and brainy: a novel synaptic architecture and a neuromorphic computing platform called SpiNNaker

Multi-memristive synapses

In the inexorable march to make computers function more like human brains (neuromorphic engineering/computing), an international team has announced its latest results in a July 10, 2018 news item on Nanowerk,

Two New Jersey Institute of Technology (NJIT) researchers, working with collaborators from the IBM Research Zurich Laboratory and the École Polytechnique Fédérale de Lausanne, have demonstrated a novel synaptic architecture that could lead to a new class of information processing systems inspired by the brain.

The findings are an important step toward building more energy-efficient computing systems that also are capable of learning and adaptation in the real world. …

A July 10, 2018 NJIT news release (also on EurekAlert) by Tracey Regan, which originated by the news item, adds more details,

The researchers, Bipin Rajendran, an associate professor of electrical and computer engineering, and S. R. Nandakumar, a graduate student in electrical engineering, have been developing brain-inspired computing systems that could be used for a wide range of big data applications.

Over the past few years, deep learning algorithms have proven to be highly successful in solving complex cognitive tasks such as controlling self-driving cars and language understanding. At the heart of these algorithms are artificial neural networks – mathematical models of the neurons and synapses of the brain – that are fed huge amounts of data so that the synaptic strengths are autonomously adjusted to learn the intrinsic features and hidden correlations in these data streams.

However, the implementation of these brain-inspired algorithms on conventional computers is highly inefficient, consuming huge amounts of power and time. This has prompted engineers to search for new materials and devices to build special-purpose computers that can incorporate the algorithms. Nanoscale memristive devices, electrical components whose conductivity depends approximately on prior signaling activity, can be used to represent the synaptic strength between the neurons in artificial neural networks.

While memristive devices could potentially lead to faster and more power-efficient computing systems, they are also plagued by several reliability issues that are common to nanoscale devices. Their efficiency stems from their ability to be programmed in an analog manner to store multiple bits of information; however, their electrical conductivities vary in a non-deterministic and non-linear fashion.

In the experiment, the team showed how multiple nanoscale memristive devices exhibiting these characteristics could nonetheless be configured to efficiently implement artificial intelligence algorithms such as deep learning. Prototype chips from IBM containing more than one million nanoscale phase-change memristive devices were used to implement a neural network for the detection of hidden patterns and correlations in time-varying signals.

“In this work, we proposed and experimentally demonstrated a scheme to obtain high learning efficiencies with nanoscale memristive devices for implementing learning algorithms,” Nandakumar says. “The central idea in our demonstration was to use several memristive devices in parallel to represent the strength of a synapse of a neural network, but only chose one of them to be updated at each step based on the neuronal activity.”

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

Neuromorphic computing with multi-memristive synapses by Irem Boybat, Manuel Le Gallo, S. R. Nandakumar, Timoleon Moraitis, Thomas Parnell, Tomas Tuma, Bipin Rajendran, Yusuf Leblebici, Abu Sebastian, & Evangelos Eleftheriou. Nature Communications volume 9, Article number: 2514 (2018) DOI: https://doi.org/10.1038/s41467-018-04933-y Published 28 June 2018

This is an open access paper.

Also they’ve got a couple of very nice introductory paragraphs which I’m including here, (from the June 28, 2018 paper in Nature Communications; Note: Links have been removed),

The human brain with less than 20 W of power consumption offers a processing capability that exceeds the petaflops mark, and thus outperforms state-of-the-art supercomputers by several orders of magnitude in terms of energy efficiency and volume. Building ultra-low-power cognitive computing systems inspired by the operating principles of the brain is a promising avenue towards achieving such efficiency. Recently, deep learning has revolutionized the field of machine learning by providing human-like performance in areas, such as computer vision, speech recognition, and complex strategic games¹. However, current hardware implementations of deep neural networks are still far from competing with biological neural systems in terms of real-time information-processing capabilities with comparable energy consumption.

One of the reasons for this inefficiency is that most neural networks are implemented on computing systems based on the conventional von Neumann architecture with separate memory and processing units. There are a few attempts to build custom neuromorphic hardware that is optimized to implement neural algorithms^2,3,4,5. However, as these custom systems are typically based on conventional silicon complementary metal oxide semiconductor (CMOS) circuitry, the area efficiency of such hardware implementations will remain relatively low, especially if in situ learning and non-volatile synaptic behavior have to be incorporated. Recently, a new class of nanoscale devices has shown promise for realizing the synaptic dynamics in a compact and power-efficient manner. These memristive devices store information in their resistance/conductance states and exhibit conductivity modulation based on the programming history^6,7,8,9. The central idea in building cognitive hardware based on memristive devices is to store the synaptic weights as their conductance states and to perform the associated computational tasks in place.

The two essential synaptic attributes that need to be emulated by memristive devices are the synaptic efficacy and plasticity. …

It gets more complicated from there.

Now onto the next bit.

SpiNNaker

At a guess, those capitalized N’s are meant to indicate ‘neural networks’. As best I can determine, SpiNNaker is not based on the memristor. Moving on, a July 11, 2018 news item on phys.org announces work from a team examining how neuromorphic hardware and neuromorphic software work together,

A computer built to mimic the brain’s neural networks produces similar results to that of the best brain-simulation supercomputer software currently used for neural-signaling research, finds a new study published in the open-access journal Frontiers in Neuroscience. Tested for accuracy, speed and energy efficiency, this custom-built computer named SpiNNaker, has the potential to overcome the speed and power consumption problems of conventional supercomputers. The aim is to advance our knowledge of neural processing in the brain, to include learning and disorders such as epilepsy and Alzheimer’s disease.

A July 11, 2018 Frontiers Publishing news release on EurekAlert, which originated the news item, expands on the latest work,

“SpiNNaker can support detailed biological models of the cortex–the outer layer of the brain that receives and processes information from the senses–delivering results very similar to those from an equivalent supercomputer software simulation,” says Dr. Sacha van Albada, lead author of this study and leader of the Theoretical Neuroanatomy group at the Jülich Research Centre, Germany. “The ability to run large-scale detailed neural networks quickly and at low power consumption will advance robotics research and facilitate studies on learning and brain disorders.”

The human brain is extremely complex, comprising 100 billion interconnected brain cells. We understand how individual neurons and their components behave and communicate with each other and on the larger scale, which areas of the brain are used for sensory perception, action and cognition. However, we know less about the translation of neural activity into behavior, such as turning thought into muscle movement.

Supercomputer software has helped by simulating the exchange of signals between neurons, but even the best software run on the fastest supercomputers to date can only simulate 1% of the human brain.

“It is presently unclear which computer architecture is best suited to study whole-brain networks efficiently. The European Human Brain Project and Jülich Research Centre have performed extensive research to identify the best strategy for this highly complex problem. Today’s supercomputers require several minutes to simulate one second of real time, so studies on processes like learning, which take hours and days in real time are currently out of reach.” explains Professor Markus Diesmann, co-author, head of the Computational and Systems Neuroscience department at the Jülich Research Centre.

He continues, “There is a huge gap between the energy consumption of the brain and today’s supercomputers. Neuromorphic (brain-inspired) computing allows us to investigate how close we can get to the energy efficiency of the brain using electronics.”

Developed over the past 15 years and based on the structure and function of the human brain, SpiNNaker — part of the Neuromorphic Computing Platform of the Human Brain Project — is a custom-built computer composed of half a million of simple computing elements controlled by its own software. The researchers compared the accuracy, speed and energy efficiency of SpiNNaker with that of NEST–a specialist supercomputer software currently in use for brain neuron-signaling research.

“The simulations run on NEST and SpiNNaker showed very similar results,” reports Steve Furber, co-author and Professor of Computer Engineering at the University of Manchester, UK. “This is the first time such a detailed simulation of the cortex has been run on SpiNNaker, or on any neuromorphic platform. SpiNNaker comprises 600 circuit boards incorporating over 500,000 small processors in total. The simulation described in this study used just six boards–1% of the total capability of the machine. The findings from our research will improve the software to reduce this to a single board.”

Van Albada shares her future aspirations for SpiNNaker, “We hope for increasingly large real-time simulations with these neuromorphic computing systems. In the Human Brain Project, we already work with neuroroboticists who hope to use them for robotic control.”

Before getting to the link and citation for the paper, here’s a description of SpiNNaker’s hardware from the ‘Spiking neural netowrk’ Wikipedia entry, Note: Links have been removed,

Neurogrid, built at Stanford University, is a board that can simulate spiking neural networks directly in hardware. SpiNNaker (Spiking Neural Network Architecture) [emphasis mine], designed at the University of Manchester, uses ARM processors as the building blocks of a massively parallel computing platform based on a six-layer thalamocortical model.[5]

Now for the link and citation,

Performance Comparison of the Digital Neuromorphic Hardware SpiNNaker and the Neural Network Simulation Software NEST for a Full-Scale Cortical Microcircuit Model by
Sacha J. van Albada, Andrew G. Rowley, Johanna Senk, Michael Hopkins, Maximilian Schmidt, Alan B. Stokes, David R. Lester, Markus Diesmann, and Steve B. Furber. Neurosci. 12:291. doi: 10.3389/fnins.2018.00291 Published: 23 May 2018

As noted earlier, this is an open access paper.

FrogHeart

Commentary about nanotech, science policy and communication, society, and the arts