Tag Archives: Qing Wu

‘Super-Turing AI’ uses less energy to mimic brain

Leave a reply

Neuromorphic (brainlike) engineering and neuromorphic computing being long time interests here, this March 26, 2025 new item on ScienceDaily caught my eye,

Artificial Intelligence (AI) can perform complex calculations and analyze data faster than any human, but to do so requires enormous amounts of energy. The human brain is also an incredibly powerful computer, yet it consumes very little energy.

As technology companies increasingly expand, a new approach to AI’s “thinking,” developed by researchers including Texas A&M University engineers, mimics the human brain and has the potential to revolutionize the AI industry.

…

A March 25, 2025 Texas A&M University news release (also on EurekAlert) by Lesley Henton, which originated the news item, delves further into the creation of a “Super-Turing AI,” Note: Links have been removed,

As technology companies increasingly expand, a new approach to AI’s “thinking,” developed by researchers including Texas A&M University engineers, mimics the human brain and has the potential to revolutionize the AI industry.

Dr. Suin Yi, assistant professor of electrical and computer engineering at Texas A&M’s College of Engineering, is on a team of researchers that developed “Super-Turing AI,” which operates more like the human brain. This new AI integrates certain processes instead of separating them and then migrating huge amounts of data like current systems do.

The Energy Crisis In AI

Today’s AI systems, including large language models [LLM] such as OpenAI [a company not an LLM] and ChatGPT [an LLM produced by OpenAI], require immense computing power and are housed in expansive data centers that consume vast amounts of electricity.

“These data centers are consuming power in gigawatts, whereas our brain consumes 20 watts,” Suin explained. “That’s 1 billion watts compared to just 20. Data centers that are consuming this energy are not sustainable with current computing methods. So while AI’s abilities are remarkable, the hardware and power generation needed to sustain it is still needed.”

The substantial energy demands not only escalate operational costs but also raise environmental concerns, given the carbon footprint associated with large-scale data centers. As AI becomes more integrated, addressing its sustainability becomes increasingly critical.

Emulating The Brain

Yi and team believe the key to solving this problem lies in nature — specifically, the human brain’s neural processes.

In the brain, the functions of learning and memory are not separated, they are integrated. Learning and memory rely on connections between neurons, called “synapses,” where signals are transmitted. Learning strengthens or weakens synaptic connections through a process called “synaptic plasticity,” forming new circuits and altering existing ones to store and retrieve information. 

By contrast, in current computing systems, training (how the AI is taught) and memory (data storage) happen in two separate places within the computer hardware. Super-Turing AI is revolutionary because it bridges this efficiency gap, so the computer doesn’t have to migrate enormous amounts of data from one part of its hardware to another.

“Traditional AI models rely heavily on backpropagation — a method used to adjust neural networks during training,” Yi said. “While effective, backpropagation is not biologically plausible and is computationally intensive.

“What we did in that paper is troubleshoot the biological implausibility present in prevailing machine learning algorithms,” he said. “Our team explores mechanisms like Hebbian learning and spike-timing-dependent plasticity — processes that help neurons strengthen connections in a way that mimics how real brains learn.”

Hebbian learning principles are often summarized as “cells that fire together, wire together.” This approach aligns more closely with how neurons in the brain strengthen their connections based on activity patterns. By integrating such biologically inspired mechanisms, the team aims to develop AI systems that require less computational power without compromising performance.

In a test, a circuit using these components helped a drone navigate a complex environment — without prior training — learning and adapting on the fly. This approach was faster, more efficient and used less energy than traditional AI.

Why This Matters For The Future Of AI

This research could be a game-changer for the AI industry. Companies are racing to build larger and more powerful AI models, but their ability to scale is limited by hardware and energy constraints. In some cases, new AI applications require building entire new data centers, further increasing environmental and economic costs.

Yi emphasizes that innovation in hardware is just as crucial as advancements in AI systems themselves. “Many people say AI is just a software thing, but without computing hardware, AI cannot exist,” he said.

Looking Ahead: Sustainable AI Development

Super-Turing AI represents a pivotal step toward sustainable AI development. By reimagining AI architectures to mirror the efficiency of the human brain, the industry can address both economic and environmental challenges.

Yi and his team hope that their research will lead to a new generation of AI that is both smarter and more efficient.

“Modern AI like ChatGPT is awesome, but it’s too expensive. We’re going to make sustainable AI,” Yi said. “Super-Turing AI could reshape how AI is built and used, ensuring that as it continues to advance, it does so in a way that benefits both people and the planet.”

There’s no mention of a memristor but there is a ‘synaptic resistor’, which I find puzzling. Is a synaptic resistor something different? In a search with these search terms “synaptic resistor memristor” I found this,

The term “memristive synapses” signifies the amalgamation of memristor functionality with synaptic characteristics, resulting in a novel approach to neuromorphic computing.

I’m guessing memristive synapses can also be called synaptic resistors or, at the least, are related concepts.

I pulled the definition from,

Resistive Switching Properties in Memristors for Optoelectronic Synaptic Memristors: Deposition Techniques, Key Performance Parameters, and Applications by Rajwali Khan, Naveed Ur Rehman, Shahid Iqbal, Sherzod Abdullaev, and Haila M. Aldosari. ACS Applied Electronic Materials Vol 6/ Issue 1 pp. 73–119 DOI: https://doi.org/10.1021/acsaelm.3c01323 Published December 29, 2023 Copyright © 2023 The Authors. Published by American Chemical Society. This publication is licensed under
CC-BY 4.0

Getting back to this latest work from Texas A&M University, here’s a link to and a citation for Dr. Suin Yi and his team’s paper,

HfZrO-based synaptic resistor circuit for a Super-Turing intelligent system by Jungmin Lee, Rahul Shenoy, Atharva Deo, Suin Yi, Dawei Gao, David Qiao, Mingjie Xu, Shiva Asapu, Zixuan Rong, Dhruva Nathan, Yong Hei, Dharma Paladugu, Jian-Guo Zheng, J. Joshua Yang, R. Stanley Williams, Qing Wu, and Yong Chen. Science Advances 28 Feb 2025 Vol 11, Issue 9 DOI: 10.1126/sciadv.adr2082

This paper is open access.

Notice that one of the Super Turing paper’s authors is R. Stanley Williams who ‘discovered’ the memristor in 2008. You can read his November 28, 2008 article “How We Found the Missing Memristor; The memristor—the functional equivalent of a synapse—could revolutionize circuit design” in the IEEE Spectrum online,

It’s time to stop shrinking. Moore’s Law, the semiconductor industry’s obsession with the shrinking of transistors and their commensurate steady doubling on a chip about every two years, has been the source of a 50-year technical and economic revolution. Whether this scaling paradigm lasts for five more years or 15, it will eventually come to an end. The emphasis in electronics design will have to shift to devices that are not just increasingly infinitesimal but increasingly capable.

Earlier this year, I and my colleagues at Hewlett-Packard Labs, in Palo Alto, Calif., surprised the electronics community with a fascinating candidate for such a device: the memristor. It had been theorized nearly 40 years ago, but because no one had managed to build one, it had long since become an esoteric curiosity. That all changed on 1 May [2008], when my group published the details of the memristor in Nature.

…

For anyone interested in a trip down memory road, I have a few comments from the theorist (Leon Chua) mentioned in his 2008 article in this April 13, 2010 posting (scroll down to the ‘More on memristors’ subhead).

Gilding medieaval statues with nanoscale gold sheets

Leave a reply
The altar examined is thought to have been made around 1420 in Southern Germany and for a long time stood in a mountain chapel on Alp Leiggern in the Swiss canton of Valais. Today it is on display at the Swiss National Museum (Landesmuseum Zürich). (Photo: Swiss National Museum, Landesmuseum Zürich) [ddownloaded from https://www.psi.ch/en/media/our-research/nanomaterial-from-the-middle-ages]

As amazing as the altar appears, it was hiding some even more amazing secrets. From an October 10, 2022 Paul Scherrer Institute (PSI) press release (also on EurekAlert but published October 11, 2022) by Barbara Vonarburg,

To gild sculptures in the late Middle Ages, artists often applied ultra-thin gold foil supported by a silver base layer. For the first time, scientists at the Paul Scherrer Institute [PSI] have managed to produce nanoscale 3D images of this material, known as Zwischgold. The pictures show this was a highly sophisticated mediaeval production technique and demonstrate why restoring such precious gilded artefacts is so difficult.

The samples examined at the Swiss Light Source SLS using one of the most advanced microscopy methods were unusual even for the highly experienced PSI team: minute samples of materials taken from an altar and wooden statues originating from the fifteenth century. The altar is thought to have been made around 1420 in Southern Germany and stood for a long time in a mountain chapel on Alp Leiggern in the Swiss canton of Valais. Today it is on display at the Swiss National Museum (Landesmuseum Zürich). In the middle you can see Mary cradling Baby Jesus. The material sample was taken from a fold in the Virgin Mary’s robe. The tiny samples from the other two mediaeval structures were supplied by Basel Historical Museum.

The material was used to gild the sacred figures. It is not actually gold leaf, but a special double-sided foil of gold and silver where the gold can be ultra-thin because it is supported by the silver base. This material, known as Zwischgold (part-gold) was significantly cheaper than using pure gold leaf. “Although Zwischgold was frequently used in the Middle Ages, very little was known about this material up to now,” says PSI physicist Benjamin Watts: “So we wanted to investigate the samples using 3D technology which can visualise extremely fine details.” Although other microscopy techniques had been used previously to examine Zwischgold, they only provided a 2D cross-section through the material. In other words, it was only possible to view the surface of the cut segment, rather than looking inside the material.  The scientists were also worried that cutting through it may have changed the structure of the sample. The advanced microscopy imaging method used today, ptychographic tomography, provides a 3D image of Zwischgold’s exact composition for the first time.

X-rays generate a diffraction pattern

The PSI scientists conducted their research using X-rays produced by the Swiss Light Source SLS. These produce tomographs displaying details in the nanoscale range – millionths of a millimetre, in other words. “Ptychography is a fairly sophisticated method, as there is no objective lens that forms an image directly on the detector,” Watts explains. Ptychography actually produces a diffraction pattern of the illuminated area, in other words an image with points of differing intensity. By manipulating the sample in a precisely defined manner, it is possible to generate hundreds of overlapping diffraction patterns. “We can then combine these diffraction patterns like a sort of giant Sudoku puzzle and work out what the original image looked like,” says the physicist. A set of ptychographic images taken from different directions can be combined to create a 3D tomogram.

The advantage of this method is its extremely high resolution. “We knew the thickness of the Zwischgold sample taken from Mary was of the order of hundreds of nanometres,” Watts explains. “So we had to be able to reveal even tinier details.” The scientists achieved this using ptychographic tomography, as they report in their latest article in the journal Nanoscale. “The 3D images clearly show how thinly and evenly the gold layer is over the silver base layer,” says Qing Wu, lead author of the publication. The art historian and conservation scientist completed her PhD at the University of Zurich, in collaboration with PSI and the Swiss National Museum. “Many people had assumed that technology in the Middle Ages was not particularly advanced,” Wu comments. “On the contrary: this was not the Dark Ages, but a period when metallurgy and gilding techniques were incredibly well developed.”

Secret recipe revealed

Unfortunately there are no records of how Zwischgold was produced at the time. “We reckon the artisans kept their recipe secret,” says Wu. Based on nanoscale images and documents from later epochs, however, the art historian now knows the method used in the 15th century: first the gold and the silver were hammered separately to produce thin foils, whereby the gold film had to be much thinner than the silver. Then the two metal foils were worked on together. Wu describes the process: “This required special beating tools and pouches with various inserts made of different materials into which the foils were inserted,” Wu explains. This was a fairly complicated procedure that required highly skilled specialists.

“Our investigations of Zwischgold samples showed the average thickness of the gold layer to be around 30 nanometres, while gold leaf produced in the same period and region was approximately 140 nanometres thick,” Wu explains. “This method saved on gold, which was much more expensive”. At the same time, there was also a very strict hierarchy of materials: gold leaf was used to make the halo of one figure, for example, while Zwischgold was used for the robe. Because this material has less of a sheen, the artists often used it to colour the hair or beards of their statues. “It is incredible how someone with only hand tools was able to craft such nanoscale material,” Watts says. Mediaeval artisans also benefited from a unique property of gold and silver crystals when pressed together: their morphology is preserved across the entire metal film. “A lucky coincidence of nature that ensures this technique works,” says the physicist.

Golden surface turns black

The 3D images do bring to light one drawback of using Zwischgold, however: the silver can push through the gold layer and cover it. The silver moves surprisingly quickly – even at room temperature. Within days, a thin silver coating covers the gold completely. At the surface the silver comes into contact with water and sulphur in the air, and corrodes. “This makes the gold surface of the Zwischgold turn black over time,” Watts explains. “The only thing you can do about this is to seal the surface with a varnish so the sulphur does not attack the silver and form silver sulphide.” The artisans using Zwischgold were aware of this problem from the start. They used resin, glue or other organic substances as a varnish. “But over hundreds of years this protective layer has decomposed, allowing corrosion to continue,” Wu explains.

The corrosion also encourages more and more silver to migrate to the surface, creating a gap below the Zwischgold. “We were surprised how clearly this gap under the metal layer could be seen,” says Watts. Especially in the sample taken from Mary’s robe, the Zwischgold had clearly come away from the base layer. “This gap can cause mechanical instability, and we expect that in some cases it is only the protective coating over the Zwischgold that is holding the metal foil in place,” Wu warns. This is a massive problem for the restoration of historical artefacts, as the silver sulphide has become embedded in the varnish layer or even further down. “If we remove the unsightly products of corrosion, the varnish layer will also fall away and we will lose everything,” says Wu. She hopes it will be possible in future to develop a special material that can be used to fill the gap and keep the Zwischgold attached. “Using ptychographic tomography, we could check how well such a consolidation material would perform its task,” says the art historian.

About PSI

The Paul Scherrer Institute PSI develops, builds and operates large, complex research facilities and makes them available to the national and international research community. The institute’s own key research priorities are in the fields of matter and materials, energy and environment and human health. PSI is committed to the training of future generations. Therefore about one quarter of our staff are post-docs, post-graduates or apprentices. Altogether PSI employs 2100 people, thus being the largest research institute in Switzerland. The annual budget amounts to approximately CHF 400 million. PSI is part of the ETH Domain, with the other members being the two Swiss Federal Institutes of Technology, ETH Zurich and EPFL Lausanne, as well as Eawag (Swiss Federal Institute of Aquatic Science and Technology), Empa (Swiss Federal Laboratories for Materials Science and Technology) and WSL (Swiss Federal Institute for Forest, Snow and Landscape Research). Insight into the exciting research of the PSI with changing focal points is provided 3 times a year in the publication 5232 – The Magazine of the Paul Scherrer Institute.

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

A modern look at a medieval bilayer metal leaf: nanotomography of Zwischgold by
Qing Wu, Karolina Soppa, Elisabeth Müller, Julian Müller, Michal Odstrcil, Esther Hsiao Rho Tsai, Andreas Späth, Mirko Holler, Manuel Guizar-Sicairos, Benjamin Butz, Rainer H. Fink, and Benjamin Watts. Nanoscale DOI: https://doi.org/10.1039/D2NR03367D First published: 10 Oct 2022

This paper is open access.

A new memristor circuit

Leave a reply

Apparently engineers at the University of Massachusetts at Amherst have developed a new kind of memristor. A Sept. 29, 2016 news item on Nanowerk makes the announcement (Note: A link has been removed),

Engineers at the University of Massachusetts Amherst are leading a research team that is developing a new type of nanodevice for computer microprocessors that can mimic the functioning of a biological synapse—the place where a signal passes from one nerve cell to another in the body. The work is featured in the advance online publication of Nature Materials (“Memristors with diffusive dynamics as synaptic emulators for neuromorphic computing”).

Such neuromorphic computing in which microprocessors are configured more like human brains is one of the most promising transformative computing technologies currently under study.

While it doesn’t sound different from any other memristor, that’s misleading. Do read on. A Sept. 27, 2016 University of Massachusetts at Amherst news release, which originated the news item, provides more detail about the researchers and the work,

J. Joshua Yang and Qiangfei Xia are professors in the electrical and computer engineering department in the UMass Amherst College of Engineering. Yang describes the research as part of collaborative work on a new type of memristive device.

Memristive devices are electrical resistance switches that can alter their resistance based on the history of applied voltage and current. These devices can store and process information and offer several key performance characteristics that exceed conventional integrated circuit technology.

“Memristors have become a leading candidate to enable neuromorphic computing by reproducing the functions in biological synapses and neurons in a neural network system, while providing advantages in energy and size,” the researchers say.

Neuromorphic computing—meaning microprocessors configured more like human brains than like traditional computer chips—is one of the most promising transformative computing technologies currently under intensive study. Xia says, “This work opens a new avenue of neuromorphic computing hardware based on memristors.”

They say that most previous work in this field with memristors has not implemented diffusive dynamics without using large standard technology found in integrated circuits commonly used in microprocessors, microcontrollers, static random access memory and other digital logic circuits.

The researchers say they proposed and demonstrated a bio-inspired solution to the diffusive dynamics that is fundamentally different from the standard technology for integrated circuits while sharing great similarities with synapses. They say, “Specifically, we developed a diffusive-type memristor where diffusion of atoms offers a similar dynamics [?] and the needed time-scales as its bio-counterpart, leading to a more faithful emulation of actual synapses, i.e., a true synaptic emulator.”

The researchers say, “The results here provide an encouraging pathway toward synaptic emulation using diffusive memristors for neuromorphic computing.”

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

Memristors with diffusive dynamics as synaptic emulators for neuromorphic computing by Zhongrui Wang, Saumil Joshi, Sergey E. Savel’ev, Hao Jiang, Rivu Midya, Peng Lin, Miao Hu, Ning Ge, John Paul Strachan, Zhiyong Li, Qing Wu, Mark Barnell, Geng-Lin Li, Huolin L. Xin, R. Stanley Williams [emphasis mine], Qiangfei Xia, & J. Joshua Yang. Nature Materials (2016) doi:10.1038/nmat4756 Published online 26 September 2016

This paper is behind a paywall.

I’ve emphasized R. Stanley Williams’ name as he was the lead researcher on the HP Labs team that proved Leon Chua’s 1971 theory about the memristor and exerted engineering control of the memristor in 2008. (Bernard Widrow, in the 1960s,  predicted and proved the existence of something he termed a ‘memistor’. Chua arrived at his ‘memristor’ theory independently.)

Austin Silver in a Sept. 29, 2016 posting on The Human OS blog (on the IEEE [Institute of Electrical and Electronics Engineers] website) delves into this latest memristor research (Note: Links have been removed),

In research published in Nature Materials on 26 September [2016], Yang and his team mimicked a crucial underlying component of how synaptic connections get stronger or weaker: the flow of calcium.

The movement of calcium into or out of the neuronal membrane, neuroscientists have found, directly affects the connection. Chemical processes move the calcium in and out— triggering a long-term change in the synapses’ strength. 2015 research in ACS NanoLetters and Advanced Functional Materials discovered that types of memristors can simulate some of the calcium behavior, but not all.

In the new research, Yang combined two types of memristors in series to create an artificial synapse. The hybrid device more closely mimics biological synapse behavior—the calcium flow in particular, Yang says.

The new memristor used–called a diffusive memristor because atoms in the resistive material move even without an applied voltage when the device is in the high resistance state—was a dielectic film sandwiched between Pt [platinum] or Au [gold] electrodes. The film contained Ag [silver] nanoparticles, which would play the role of calcium in the experiments.

By tracking the movement of the silver nanoparticles inside the diffusive memristor, the researchers noticed a striking similarity to how calcium functions in biological systems.

A voltage pulse to the hybrid device drove silver into the gap between the diffusive memristor’s two electrodes–creating a filament bridge. After the pulse died away, the filament started to break and the silver moved back— resistance increased.

Like the case with calcium, a force made silver go in and a force made silver go out.

To complete the artificial synapse, the researchers connected the diffusive memristor in series to another type of memristor that had been studied before.

When presented with a sequence of voltage pulses with particular timing, the artificial synapse showed the kind of long-term strengthening behavior a real synapse would, according to the researchers. “We think it is sort of a real emulation, rather than simulation because they have the physical similarity,” Yang says.

I was glad to find some additional technical detail about this new memristor and to find the Human OS blog, which is new to me and according to its home page is a “biomedical blog, featuring the wearable sensors, big data analytics, and implanted devices that enable new ventures in personalized medicine.”