Chips At Stake

Integrating dimensions to get more out of Moore's Law and advance electronics

Jan 2024, phys.org

https://techxplore.com/news/2024-01-dimensions-law-advance-electronics.html

Moore's Law - the number of transistors on a chip will double every two years (but a limit exists)

More Moore - vertically stacking multiple layers of semiconductor devices to beat the limit (3D integration)

More than Moore - transistors made from 2D materials (using 2D materials, not just 2D-like layers of 3D materials)

Monolithic Moore - monolithic 3D integration uses one brick of 2D things, instead of layers of 2D things, so there's no connections between the layers, so it saves space 

via Penn State: Darsith Jayachandran et al, Three-dimensional integration of two-dimensional field-effect transistors, Nature (2024). DOI: 10.1038/s41586-023-06860-5

https://pubmed.ncbi.nlm.nih.gov/38200300/

Image credit: AI Art - Taoist Octopus Tesseract Fractal Hypercube - 2024

Emulating neurodegeneration and aging in artificial intelligence systems

Apr 2024, phys.org

https://techxplore.com/news/2024-04-emulating-neurodegeneration-aging-artificial-intelligence.html

"We used IQ tests performed by large language models (LLMs) and, more specifically, the LLaMA 2, to introduce the concept of 'neural erosion. This deliberate erosion involves ablating synapses or neurons or adding Gaussian noise during or after training, resulting in a controlled decline in the LLMs' performance."

The researchers found that when they deliberately ablated (i.e., removed) some of the artificial synapses or neurons of the LLaMA 2 model, its performance on IQ tests declined, following a particular pattern.

"The LLM loses abstract thinking abilities, followed by mathematical degradation, and ultimately, a loss in linguistic ability, responding to prompts incoherently. ... We are now conducting further tests to better understand this observed pattern."

Interestingly, this 'neuro-erosion' pattern is aligned with the neurodegeneration patterns observed in humans.

via University of California Irvine: Antonios Alexos et al, Neural Erosion: Emulating Controlled Neurodegeneration and Aging in AI Systems, arXiv (2024). DOI: 10.48550/arxiv.2403.10596

https://dx.doi.org/10.48550/arxiv.2403.10596

Post Script: A new kind of chip called an NPU for neural processing unit, is like the new GPU, which was the new CPU:

Your current PC probably doesn’t have an AI processor, but your next one might

Feb 2024, Ars Technica

https://arstechnica.com/ai/2024/02/your-current-pc-probably-doesnt-have-an-ai-processor-but-your-next-one-might/

The comments in this article are the most clearly examined examples of what NPUs will do than any other thing I've read, and considering that the forum of Ars is known for the technical proficiency of its members: 

• From the article, Intel Senior Director of Technical Marketing Robert Hallock  - "Camera segmentation, this whole background blurring thing... moving that to the NPU saves about 30 to 50 percent power versus running it elsewhere."
• **RTS** - It is pretty shocking looking at powermetrics on macOS to see how rare the Neural Engine is even powered as most applications (even AI exclusive applications like stable diffusion/LLM frontends) just run any AI work they need on the CPU or GPU.
• AlicePlaysWithRockets - I use it daily in Final Cut Pro and meeting camera effects.
• LlamaDragon - NPUs can be used by Photoprism to speed up importing of images (it tries to ID pics with dogs or cars or "cooking" and so on) or by something like Frigate (camera monitoring) to do similar ID-ing in real time, it might be useful.
• AmanoJyaku Ars Praefectus - [why] It has to do with CPUs being general purpose, and therefore able to do anything, vs. hardware accelerators that do specific things. Hardware accelerators can't do most of the things CPUs can do, but the things they can do can be done much faster than CPUs do them. In turn, this makes them more power efficient. The trend started with graphics cards, continued with sound cards and network cards, and have since grown to include other devices, some of which accelerate things as minute as image decoding. ... Theoretically, the most efficient device is one that has accelerators for everything you do on a device. However, there will always be new things, so CPUs are unlikely to be eliminated. As for what makes general-purpose CPUs different from task-specific accelerators, forum posts can't easily sum this up. That's a topic explained in college courses, and the basis for entire careers. 
• autostop - The Neural Engine on the Apple chips is what makes "Live Text" (universal OCR) work on the Macintosh and iPhone. (You can open a scanned PDF, or a screen grab, and the mouse pointer turns into an I-beam and you can highlight and copy just as if it were text in Word or something.)
• Siosphere - Users who view photos a lot on their laptop/desktop, it could be used for on-device people finding, or computing interoplation for zooming in/out of a photo so it is smoother/faster. ... It could be used for generic type-ahead suggestions, which if they suck is super annoying, but when they work well are a very useful timesaver. (When they actually suggest what I was actually going to write, I'm really pleased, when they suggest something else they are annoying, but they are getting better). ... Better searching capabilities by understanding more of what you are searching for on your computer, being smarter about intent, are you searching for a specific file by name, type of file, application, etc. ... Generalized automation of repetitive tasks, things you might currently script yourself could be automatically found and setup (again, this is annoying if it doesn't do what you want, but I'm just saying best case it automates in the exact way you are wanting). ... Faster windows hello (if you use that), better camera processing for video calls, better workload predicting for changing the task priority to save battery on laptops, or give CPU power more to the things that need it. ... There are a lot of subtle ways that a dedicated AI chip could be used, and there will be so many more when it is just an available resource any developers can tap into
• Isaacc7 - NPUs can be used to accelerate many tasks like voice recognition, camera processing, live subtitles, etc. 
• Toastr - Also things like on-device automatic image tagging of faces for your photo library, which is a big plus for me who wants to organize my family photo library but doesn't like the privacy implications of cloud photo services like Google Photos. ... Or, for another example:
• The self-hosted Frigate NVR software supports object detection via CPU, GPU, or NPU. Sure, you could run it on an RTX 4080, but you could also just throw a $25 Coral Accelerator in a mini PC and get plenty of performance but with a device that draws just a few watts.
• longhornchris04 - Side note, while GPUs can do the highly parallel low precision computing, they are designed for graphics which generally require a higher level of precision. So yes, GPUs can do the work, and do it quite easily, but they are often overkill for the job and thus are less efficient at it. 
• or just anandtech: https://www.anandtech.com/show/20046/intel-unveils-meteor-lake-architecture-intel-4-heralds-the-disaggregated-future-of-mobile-cpus/4

Everything Is A Computer If You Try Hard Enough

You walk into a room, and it's empty, but you enter, and without hesitation, you sit down into mid air, and before you land, a chair materializes itself to catch you. The actual matter, the particles that make up the chair, have been engineered to become chairs. They have "chair" written not into their DNA, but into their fundamental physics. They have no power source because they take their energy from light waves, sound waves, vibrations, even gradients like in between high temperatures and lows, or saltwater and fresh, or whatever that means, since it could be an information gradient (could it?). They have no battery because they don't use more than they need in real time. Every particle of this special form of matter is a computer, with wireless communication, with sensors, all of it built into the physics of the particles themselves. It will know you're in the room, who you are and what you want. And that's why all you have to do is want to sit, and the chair materializes. It's still too hard to explain what it means when everything is a computer, but here is an introduction:  

Photonic chip that 'fits together like Lego' opens door to semiconductor industry

Dec 2023, phys.org

https://phys.org/news/2023-12-photonic-chip-lego-door-semiconductor.html

Chiplets - The chip is built using an emerging technology in silicon photonics that allows the integration of diverse systems on semiconductors less than 5 millimeters wide; it's like fitting together Lego building blocks, where new materials are integrated through advanced packaging of components, using electronic "chiplets."

via University of Sydney Nano Institute:  Matthew Garrett et al, Integrated microwave photonic notch filter using a heterogeneously integrated Brillouin and active-silicon photonic circuit, Nature Communications (2023). DOI: 10.1038/s41467-023-43404-x

https://dx.doi.org/10.1038/s41467-023-43404-x

Unexpected AI fingers image credit: AI Art - AI Fingers at Work on a Circuitboard - 2023

Study suggests that physical processes can have hidden neural network-like abilities

Jan 2024, phys.org

https://phys.org/news/2024-01-physical-hidden-neural-network-abilities.html

Natural molecular processes can do complex calculations that rival a simple neural network. On the face of it, the initial steps in the act of freezing - called 'nucleation' in physics - do not resemble 'thinking'. But the new study shows that the act of freezing can "recognize" subtly different chemical combinations - e.g., the smell of oatmeal raisin cookies versus chocolate chip - and build different molecular structures in response.

The work points at a new view of computation that does not involve designing circuits, but rather designing what physicists call a phase diagram. For example, for water, a phase diagram might describe the temperature and pressure conditions in which liquid water will freeze or boil, which are 'muscle'-like material properties. But this work shows that the phase diagram can also encode 'thinking' in addition to 'doing,' when scaled up to complex systems with many different kinds of components.

via University of Chicago, California Institute of Technology, and Maynooth University: Constantine Glen Evans et al, Pattern recognition in the nucleation kinetics of non-equilibrium self-assembly, Nature (2024). DOI: 10.1038/s41586-023-06890-z

https://dx.doi.org/10.1038/s41586-023-06890-z

 

International research team develops new hardware for neuromorphic computing

Feb 2024, phys.org

https://techxplore.com/news/2024-02-international-team-hardware-neuromorphic.html

In the eye, the visual information is pre-processed by hundreds of millions of the retina's photoreceptors and converted into electrical signals that are transmitted by the optic nerve to the brain. This process greatly reduces the amount of data processed in the brain by the visual cortex.

(The nose is far, far crazier in what it does, by the way)

Inspired by eyesight, this on-chip phonon-magnon reservoir for neuromorphic computing maps input signals into a multidimensional reservoir space. The reservoir is made of acoustic waves (phonons) and spin waves (magnons), and is not trained but only expedites recognition by a simplified artificial neural network, resulting in enormous reduction of computational resources and training time.

(Wait until they get inspired by the nose)

via Technische Universität Dortmund, Loughborough University, V. E. Lashkaryov Institute of Semiconductor Physics in Kyiv, University of Nottingham: Dmytro D. Yaremkevich et al, On-chip phonon-magnon reservoir for neuromorphic computing, Nature Communications (2023). DOI: 10.1038/s41467-023-43891-y

https://dx.doi.org/10.1038/s41467-023-43891-y

A low-cost system to collect EEG measurements during VR experiences

Feb 2024, phys.org

https://medicalxpress.com/news/2024-02-eeg-vr.html

Everything gets it's own chip, just like this

NeuroVista, the new system proposed by the researchers, utilizes KS1092, a cost-effective biological potential measurement chip. The prototype of the device created by the researchers is comprised of this chip, along with a set of electrodes, and a lithium battery.

via South China University of Technology: Zhiyuan Yu et al, A low-cost, wireless, 4-channel EEG measurement system used in virtual reality environments, HardwareX (2024). DOI: 10.1016/j.ohx.2024.e00507.

https://dx.doi.org/10.1016/j.ohx.2024.e00507

 

Giant leap toward neuromorphic devices: High-performance spin-wave reservoir computing

Mar 2024, phys.org

https://phys.org/news/2024-03-giant-neuromorphic-devices-high-reservoir.html

It's a high-performance spin wave reservoir computing that uses spintronics. It works with a randomly generated network called the "reservoir" which enables the memorization of past input information and its nonlinear transformation, allowing physical systems to perform tasks for sequential data.

via Tohoku University Advanced Institute for Materials Research: Satoshi Iihama et al, Universal scaling between wave speed and size enables nanoscale high-performance reservoir computing based on propagating spin-waves, npj Spintronics (2024). DOI: 10.1038/s44306-024-00008-5

https://dx.doi.org/10.1038/s44306-024-00008-5

Gamma Jacking

New approach puts brain scans on the witness stand in trademark disputes

Feb 2023, phys.org

https://phys.org/news/2023-02-approach-brain-scans-witness-trademark.html

Talking to brains instead of people. (Something about this reminds me of the perennial line by business owners who resist organized worker unions because they prefer "a direct relationship with their workers".)

The standard according to trademark law is whether a "reasonable person" would find two trademarks similar, but it doesn't define what similar means. "Asking the brain, not a person, could reduce -- if not eliminate -- inconsistencies." (They use fMRI).

via University of California - Berkeley: Zhihao Zhang et al, From scanner to court: A neuroscientifically informed "reasonable person" test of trademark infringement, Science Advances (2023). DOI: 10.1126/sciadv.abo1095

https://pubmed.ncbi.nlm.nih.gov/36753556/

Image credit: Gamma Ray Burst GRB 221009A nicknamed the BOAT - Brightest Of All Time - NASA Swift - 2023

Brain-inspired computing system based on skyrmions 'reads' handwriting

Feb 2023, phys.org

https://techxplore.com/news/2023-02-brain-inspired-based-skyrmions.html

The researchers trained the device using more than 13,000 images of handwritten digits from 0 to 9. They converted the images into magnetic input signals, and tuned the device so that the output voltage signals accurately represented the correct digit.

(Skyrmions are miniature magnetic whirlpools.)

via RIKEN Center for Emergent Matter Science: Tomoyuki Yokouchi et al, Pattern recognition with neuromorphic computing using magnetic field–induced dynamics of skyrmions, Science Advances (2022). DOI: 10.1126/sciadv.abq5652

https://dx.doi.org/10.1126/sciadv.abq5652

Electrodes grown in the brain: Paving the way for future therapies for neurological disorders

Feb 2023, phys.org

https://phys.org/news/2023-02-electrodes-grown-brain-paving-future.html

Like it's not a big deal:

Successfully grown electrodes in living tissue using the body's molecules as triggers.

"For several decades, we have tried to create electronics that mimic biology. Now we let biology create the electronics for us,"

The body's endogenous molecules are enough to trigger the formation of electrodes. There is no need for genetic modification or external signals, such as light or electrical energy, which has been necessary in previous experiments. The Swedish researchers are the first in the world to succeed in this.

via Linköping, Lund and Gothenburg Universities in Sweden: Xenofon Strakosas et al, Metabolite-induced in vivo fabrication of substrate-free organic bioelectronics, Science (2023). DOI: 10.1126/science.adc9998

https://dx.doi.org/10.1126/science.adc9998

Post Script:

A place to exercise your brain? Introducing mental health gyms

Feb 2023, BBC News

https://www.bbc.com/news/world-us-canada-64042267

At Inception, Mr McCullar has designed boot camps and circuit training featuring equipment to help the brain relax: infrared saunas, zero-gravity chairs, flotation therapy tanks and neurofeedback therapy.

Post Post Script

Study examines how our native language shapes our brain wiring

Mar 2023, phys.org

https://medicalxpress.com/news/2023-03-native-language-brain-wiring.html

With the help of magnetic resonance tomography, they looked deep into the brains of native German and Arabic speakers and discovered differences in the wiring of the language regions in the brain.

"Arabic native speakers showed a stronger connectivity between the left and right hemispheres than German native speakers," explained Alfred Anwander, last author of the study that was recently published in the journal NeuroImage. "This strengthening was also found between semantic language regions and may be related to the relatively complex semantic and phonological processing in Arabic."

As the researchers discovered, native German speakers showed stronger connectivity in the left hemisphere language network. They argue that their findings may be related to the complex syntactic processing of German, which is due to the free word order and greater dependency distance of sentence elements.

via Max Planck Institute for Human Cognitive and Brain Sciences in Leipzig: Xuehu Wei et al, Native language differences in the structural connectome of the human brain, NeuroImage (2023). DOI: 10.1016/j.neuroimage.2023.119955

https://dx.doi.org/10.1016/j.neuroimage.2023.119955

Ancient Mathematics is the New Mathematics

A tool to detect higher-order phenomena in real-world data

Jan 2023, phys.org

https://phys.org/news/2023-01-tool-higher-order-phenomena-real-world.html

Researchers' multivariate time series method was able to detect oscillations between chaotic and synchronized neural interactions occurring in a brain at rest, between periods of financial stability and crisis. In the epidemiological example, or interactions between the spread of different diseases, like flu and pertussis.

"We are able to use ancient mathematics in new ways thanks to modern computing power, and access to big data. We are creating a new mathematics."

via Ecole Polytechnique Federale de Lausanne, Neuro-X Institute, Austria's Central European University and Italy's CENTAI Institute: Andrea Santoro et al, Higher-order organization of multivariate time series, Nature Physics (2023). DOI: 10.1038/s41567-022-01852-0

https://dx.doi.org/10.1038/s41567-022-01852-0

Image credit: Anatoly Fomenko

Harnessing incoherence to make sense of real-world networks

Mar 2023, phys.org

https://phys.org/news/2023-03-harnessing-incoherence-real-world-networks.html

Mapping the hierarchies and also the incoherence within a system will enable us to predict the system's strong and weak points.

Most real-world systems are neither perfectly coherent nor completely incoherent, but lie somewhere in between. In a food web, for instance, this might occur because of omnivorous animals that will eat both plants and other animals.

It was possible to use this trophic incoherence to estimate the point at which a network becomes strongly connected. They demonstrated that the method works for any type of network, including those of neurons, people, species, metabolites, genes and words...

"This modeling approach could be used to disrupt networks as well, because the points at which connectivity becomes strong can be targeted. Neurologists, for example, might find new ways to treat epilepsy by pinpointing specific connections responsible for maintaining seizures."

via University of Birmingham: Niall Rodgers et al, Strong connectivity in real directed networks, Proceedings of the National Academy of Sciences (2023). DOI: 10.1073/pnas.2215752120

https://dx.doi.org/10.1073/pnas.2215752120

In the Future Matter Is Intelligent

Floppy or not: AI predicts properties of complex metamaterials

Nov 2022, phys.org

https://techxplore.com/news/2022-11-floppy-ai-properties-complex-metamaterials.html

With infinite options, infinite intelligence?

Also words:

Artificial materials - These are engineered materials whose properties are determined by their geometrical structure rather than their chemical composition [like origami].

I must have missed the part when we started calling them artificial materials, I thought they were all metamaterials.

Designing these materials is a combinatorial problem, which means it's hard. You can't really predict what will happen, you just have to do it. But artificial intelligence can do it virtually, all day, and find the ones that work. 

via University of Amsterdam: Ryan van Mastrigt et al, Machine Learning of Implicit Combinatorial Rules in Mechanical Metamaterials, Physical Review Letters (2022). DOI: 10.1103/PhysRevLett.129.198003

https://dx.doi.org/10.1103/PhysRevLett.129.198003

Image credit: AI Art - Fibonacci Alien Library 2 - 2022

Clear window coating could cool buildings without using energy

Nov 2022, phys.org

https://techxplore.com/news/2022-11-window-coating-cool-energy.html

A "transparent radiative cooler" could lower the temperature inside buildings, without expending a single watt of energy. 

The team constructed computer models of TRCs consisting of alternating thin layers of common materials like silicon dioxide, silicon nitride, aluminum oxide or titanium dioxide on a glass base, topped with a film of polydimethylsiloxane. They optimized the type, order and combination of layers using an iterative approach guided by machine learning and quantum computing, which stores data using subatomic particles. 

Cooling accounts for about 15% of global energy consumption; this thing can potentially reduce cooling energy consumption by 31% compared with conventional windows.

via Notre Dame: High-Performance Transparent Radiative Cooler Designed by Quantum Computing, ACS Energy Letters (2022). DOI: 10.1021/acsenergylett.2c01969

https://dx.doi.org/10.1021/acsenergylett.2c01969

Photovoltaic windows unlock goal of increased energy efficiency for skyscrapers

Nov 2022, phys.org

https://techxplore.com/news/2022-11-photovoltaic-windows-goal-energy-efficiency.html

Energy use climbs when a building has more windows than wall space, yet larger floor-to-floor height coupled with PV glazing reduces building energy use. 

via National Renewable Energy Laboratory: Vincent M. Wheeler et al, Photovoltaic windows cut energy use and CO2 emissions by 40% in highly glazed buildings, One Earth (2022). DOI: 10.1016/j.oneear.2022.10.014

https://dx.doi.org/10.1016/j.oneear.2022.10.014

New study suggests mobile data collected while traveling over bridges could help evaluate their integrity

Nov 2022, phys.org

https://techxplore.com/news/2022-11-mobile-bridges.html

I can see a future where we intercept wifi signals from building occupants, and measure their interactions to determine not only the building materials getting hit by the wifi waves, but their changes over time:

"Information about structural health of bridges can be extracted from smartphone-collected accelerometer data"

via MIT: Thomas Matarazzo, Crowdsourcing bridge dynamic monitoring with smartphone vehicle trips, Communications Engineering (2022). DOI: 10.1038/s44172-022-00025-4.

https://dx.doi.org/10.1038/s44172-022-00025-4

AI Art - Fibonacci Alien Library 1 - 2022

Centimeter-scale multicolor printing with a pixelated optical cavity

Nov 2022, phys.org

https://phys.org/news/2022-11-centimeter-scale-multicolor-pixelated-optical-cavity.html

"pixelated optical cavity"

The colorful image with multiple color components is first converted to a predefined grayscale pattern and then engraved on the photoresist layer by controlling the exposure dose during the grayscale laser writing process.

Pixelated photoresist spacer layers are sandwiched by two semitransparent sliver thin films to form the Fabry–Perot cavities (pixelated optical cavities). The transmission color can be continuously tuned in the visible spectral regime by finely controlling the thickness of the photoresist layer. 

via Southern University of Science and Technology in Shenzhen: Yu Chen et al, Centimeter scale color printing with grayscale lithography, Advanced Photonics Nexus (2022). DOI: 10.1117/1.APN.1.2.026002

https://dx.doi.org/10.1117/1.APN.1.2.026002

Team creates crystals that generate electricity from heat

Nov 2022, phys.org

https://phys.org/news/2022-11-team-crystals-electricity.html

This novel synthetic material is composed of copper, manganese, germanium, and sulfur, and it is produced by simple ball-milling and then heating to 600 degrees Celsius. 

It's called a "thermoelectric material" because it converts heat to electricity. 

via Normandie University: V. Pavan Kumar et al, Engineering Transport Properties in Interconnected Enargite‐Stannite Type Cu 2+ x Mn 1− x GeS 4 Nanocomposites, Angewandte Chemie International Edition (2022). DOI: 10.1002/anie.202210600

https://dx.doi.org/10.1002/anie.202210600

Mimicking life: A breakthrough in non-living materials

Nov 2022, phys.org

https://phys.org/news/2022-11-mimicking-life-breakthrough-non-living-materials.html

Artificial Life - Ok they're calling them all kinds of things, now including "non-living materials", also related to soft robotics:

New process that uses fuel to control non-living materials at a specified rate, similar to what living cells do

"Ultimately you'd want a robot to be able to control itself. You can program our cycle into a particle in advance, then leave it alone, and it performs its function independently as soon as it encounters a signal to do so."

Particles man.

via Delft University of Technology: Benjamin Klemm et al, Temporally programmed polymer—solvent interactions using a chemical reaction network, Nature Communications (2022). DOI: 10.1038/s41467-022-33810-y

https://dx.doi.org/10.1038/s41467-022-33810-y

Discovery reveals 'brain-like computing' at molecular level is possible

Nov 2022, phys.org

https://phys.org/news/2022-11-discovery-reveals-brain-like-molecular.html

Brains all the way down:

"Intelligent molecular materials"

Disruptive new alternative to conventional silicon-based digital switches that can only ever be either on or off. It displays all the mathematical logic functions necessary for deep learning.

"The community has long known that silicon technology works completely differently to how our brains work and so we used new types of electronic materials based on soft molecules to emulate brain-like computing networks."

via  University of Limerick's Bernal Institute: Enrique del Barco, Dynamic molecular switches with hysteretic negative differential conductance emulating synaptic behaviour, Nature Materials (2022). DOI: 10.1038/s41563-022-01402-2

https://dx.doi.org/10.1038/s41563-022-01402-

AI Art - Mobius in an Escher Room with Penrose Triangles - 2022

Self-assembled nanoscale architectures could feature improved electronic, optical, and mechanical properties

Nov 2022, phys.org

https://phys.org/news/2022-11-self-assembled-nanoscale-architectures-feature-electronic.html

Internet of Everything 

"Self-assembly is a really beautiful way to make structures," Yager said. "You design the molecules, and the molecules spontaneously organize into the desired structure."

via Department of Energy's Brookhaven National Laboratory's Center for Functional Nanomaterials: Sebastian T. Russell et al, Priming self-assembly pathways by stacking block copolymers, Nature Communications (2022). DOI: 10.1038/s41467-022-34729-0

https://dx.doi.org/10.1038/s41467-022-34729-0

Breakthrough algorithm expands the exploration space for materials by orders of magnitude

Nov 2022, phys.org

https://techxplore.com/news/2022-11-breakthrough-algorithm-exploration-space-materials.html

Algorithm that predicts the structure and dynamic properties of any material—whether existing or new—almost instantaneously.

It's called M3GNet and it was used to develop matterverse.ai, a database of more than 31 million yet-to-be-synthesized materials with properties predicted by machine learning algorithms. 

via University of California San Diego: Chi Chen, A universal graph deep learning interatomic potential for the periodic table, Nature Computational Science (2022). DOI: 10.1038/s43588-022-00349-3

https://dx.doi.org/10.1038/s43588-022-00349-32

Kirigami technique hints at promising outcomes for breast reconstruction

Dec 2022, phys.org

https://medicalxpress.com/news/2022-12-kirigami-technique-hints-outcomes-breast.html

Kirigami boobs

via University of Pennsylvania: Young‐Joo Lee et al, Natural Shaping of Acellular Dermal Matrices for Implant‐Based Breast Reconstruction via Expansile Kirigami, Advanced Materials (2022). DOI: 10.1002/adma.202208088

https://dx.doi.org/10.1002/adma.202208088

Ubiquitous Intelligence

Solving the 'big problems' via algorithms enhanced by 2D materials

Jan 2022, phys.org

https://techxplore.com/news/2022-01-big-problems-algorithms-2d-materials.html

Imagining the future is hard; you almost never get it right. You just can't see what's not there.  But in this case, a glimpse reveals itself -- every "computer" will be designed for a specific algorithm. There won't be a "new" way of making a computer, or of making a one-size-fits-all computer that's faster or better. The very idea of a one-size-fits-all computer is what makes it hard for us to see the future. Before the electric guitar was invented, the acoustic guitar didn't exist, it was just called a guitar. (Like smartphones and dumbphones.)

Eventually, you won't have an advanced computer that can run different algorithms better, instead the computer and the algorithm will be one, and therefore there will be as many types of computers as there are algorithms. Like the Cambrian explosion, but different. 

The "combinatorial optimization problem" they're solving here is also referred to as the "traveling salesman problem", or the "design an optimal transit system based on the terrain of the region, distribution of the population, existing routes, etc.", or the "use a living slime mold computer to design an optimal transit system" problem. 

The reason it's so hard for our current algorithms to do the optimization problem is because computers as we know them today are still based on a design from the 1940's. The problem isn't so much because they're old, it's because we don't work with data the same way we used to. We have a lot more data than we used to, and integrating it all at the same time is hard for today's computers.

I like to think of it simply as a problem where your database has as many columns as it does rows. This is what happens when you try to categorize smells based on the names we call them. On one axis you have all the smellable molecules there are (veritably infinite), and on the other, you have all the attributes you can give to any one of the molecules (physical dimensions, descriptions, names, autobiographical physiodata that your body associates with the molecule, which is also veritably infinite). You would then have a database, a spreadsheet of infinite cells. It's hard to work with something that big. 

But we don't have to do things like that anymore. Instead, we can use slime mold, or we can design "new computers" that combine information storage and computing into the same thing. This sounds a lot like a neuromorphic computer, by the way.

via Pennsylvania State University: Amritanand Sebastian et al, An Annealing Accelerator for Ising Spin Systems Based on In‐Memory Complementary 2D FETs, Advanced Materials (2021). DOI: 10.1002/adma.202107076

http://dx.doi.org/10.1002/adma.202107076

Image credit: Flows of individuals across the Greater Boston area, Guangyu Du at Sante Fe Inst, 2021

Post Script:

And how they do it? A form of "in-memory computing" based on simulated annealing, where atoms reorganize themselves and then crystallize in the lowest energy state. Sounds a lot like 2-D metamaterials, BECs and quantum crystallography. Putting it all together. 

Notes:

Using a 'virtual slime mold' to design a subway network less prone to disruption

Feb 2022, phys.org

https://phys.org/news/2022-01-virtual-slime-mold-subway-network.html

A model, no slime needed.

via University of Toronto: Raphael Kay et al, Stepwise slime mould growth as a template for urban design, Scientific Reports (2022). DOI: 10.1038/s41598-022-05439-w

http://dx.doi.org/10.1038/s41598-022-05439-w