Aurelia Butler

Warnings have emerged about the unreliability of the metrics used to detect whether an audio perturbation designed to fool AI models can be perceived by humans. Researchers at the UPV/EHU-University of the Basque Country show that the distortion metrics used to detect intentional perturbations in audio signals are not a reliable measure of human perception, and have proposed a series of improvements. These perturbations, designed to be imperceptible, can be used to cause erroneous predictions in artificial intelligence. Distortion metrics are applied to assess how effective the methods are in generating such attacks.
Artificial intelligence (AI) is increasingly based on machine learning models, trained using large datasets. Likewise, human-computer interaction is increasingly dependent on speech communication, mainly due to the remarkable performance of machine learning models in speech recognition tasks.
However, these models can be fooled by “adversarial” examples, in other words, inputs intentionally perturbed to produce a wrong prediction without the changes being noticed by humans. “Suppose we have a model that classifies audio (e.g. voice command recognition) and we want to deceive it, in other words, generate a perturbation that maliciously prevents the model from working properly. If a signal is heard properly, a person is able to notice whether a signal says ‘yes’, for example. When we add an adversarial perturbation we will still hear ‘yes’, but the model will start to hear ‘no’, or ‘turn right’ instead of left or any other command we don’t want to execute,” explained Jon Vadillo, researcher in the UPV/EHU’s Departament of Computer Science and Artificial Intelligence.
This could have “very serious implications at the level of applying these technologies to real-world or highly sensitive problems,” added Vadillo. It remains unclear why this happens. Why would a model that behaves so intelligently suddenly stop working properly when it receives even slightly altered signals?
Deceiving the model by using an undetectable perturbation
“It is important to know whether a model or a programme has vulnerabilities,” added the researcher from the Faculty of Informatics. “Firstly, we investigate these vulnerabilities, to check that they exist, and because that is the first step in eventually fixing them.” While much research has focused on the development of new techniques for generating adversarial perturbations, less attention has been paid to the aspects that determine whether these perturbations can be perceived by humans and what these aspects are like. This issue is important, as the adversarial perturbation strategies proposed only pose a threat if the perturbations cannot be detected by humans.
This study has investigated the extent to which the distortion metrics proposed in the literature for audio adversarial examples can reliably measure the human perception of perturbations. In an experiment in which 36 people evaluated adversarial examples or audio perturbations according to various factors, the researchers showed that “the metrics that are being used by convention in the literature are not completely robust or reliable. In other words, they do not adequately represent the auditory perception of humans; they may tell you that a perturbation cannot be detected, but then when we evaluate it with humans, it turns out to be detectable. So we want to issue a warning that due to the lack of reliability of these metrics, the study of these audio attacks is not being conducted very well,” said the researcher.
In addition, the researchers have proposed a more robust evaluation method that is the outcome of the “analysis of certain properties or factors in the audio that are relevant when assessing detectability, for example, the parts of the audio in which a perturbation is most detectable.” Even so, “this problem remains open because it is very difficult to come up with a mathematical metric that is capable of modelling auditory perception. Depending on the type of audio signal, different metrics will probably be required or different factors will need to be considered. Achieving general audio metrics that are representative is a complex task,” concluded Vadillo.
Story Source:
Materials provided by University of the Basque Country. Note: Content may be edited for style and length. More

When two sheets of the carbon nanomaterial graphene are stacked together at a particular angle with respect to each other, it gives rise to some fascinating physics. For instance, when this so-called “magic-angle graphene” is cooled to near absolute zero, it suddenly becomes a superconductor, meaning it conducts electricity with zero resistance.
Now, a research team from Brown University has found a surprising new phenomenon that can arise in magic-angle graphene. In research published in the journal Science, the team showed that by inducing a phenomenon known as spin-orbit coupling, magic-angle graphene becomes a powerful ferromagnet.
“Magnetism and superconductivity are usually at opposite ends of the spectrum in condensed matter physics, and it’s rare for them to appear in the same material platform,” said Jia Li, an assistant professor of physics at Brown and senior author of the research. “Yet we’ve shown that we can create magnetism in a system that originally hosts superconductivity. This gives us a new way to study the interplay between superconductivity and magnetism, and provides exciting new possibilities for quantum science research.”
Magic-angle graphene has caused quite a stir in physics in recent years. Graphene is a two-dimensional material made of carbon atoms arranged in a honeycomb-like pattern. Single sheets of graphene are interesting on their own — displaying remarkable material strength and extremely efficient electrical conductance. But things get even more interesting when graphene sheets are stacked. Electrons begin to interact not only with other electrons within a graphene sheet, but also with those in the adjacent sheet. Changing the angle of the sheets with respect to each other changes those interactions, giving rise to interesting quantum phenomena like superconductivity.
This new research adds a new wrinkle — spin-orbit coupling — to this already interesting system. Spin-orbit coupling is a state of electron behavior in certain materials in which each electron’s spin — its tiny magnetic moment that points either up or down — becomes linked to its orbit around the atomic nucleus.
“We know that spin-orbit coupling gives rise to a wide range of interesting quantum phenomena, but it’s not normally present in magic-angle graphene,” said Jiang-Xiazi Lin, a postdoctoral researcher at Brown and the study’s lead author. “We wanted to introduce spin-orbit coupling, and then see what effect it had on the system.”
To do that, Li and his team interfaced magic-angle graphene with a block of tungsten diselenide, a material that has strong spin-orbit coupling. Aligning the stack precisely induces spin-orbit coupling in the graphene. From there, the team probed the system with external electrical currents and magnetic fields. More

A pioneering type of patented computer memory known as ULTRARAM™ has been demonstrated on silicon wafers in what is a major step towards its large-scale manufacture.
ULTRARAM™ is a novel type of memory with extraordinary properties. It combines the non-volatility of a data storage memory, like flash, with the speed, energy-efficiency and endurance of a working memory, like DRAM. To do this it utilises the unique properties of compound semiconductors, commonly used in photonic devices such as LEDS, laser diodes and infrared detectors, but not in digital electronics, which is the preserve of silicon.
Initially patented in the US, further patents on the technology are currently being progressed in key technology markets around the world.
Now, in a collaboration between the Physics and Engineering Departments at Lancaster University and the Department of Physics at Warwick, ULTRARAM™ has been implemented on silicon wafers for the very first time.
Professor Manus Hayne of the Department of Physics at Lancaster, who leads the work said, “ULTRARAM™ on silicon is a huge advance for our research, overcoming very significant materials challenges of large crystalline lattice mismatch, the change from elemental to compound semiconductor and differences in thermal contraction.”
Digital electronics, which is the core of all gadgetry from smart watches and smart phones through to personal computers and datacentres, uses processor and memory chips made from the semiconductor element silicon.
Due to the maturity of the silicon chip-making industry and the multi-billion dollar cost of building chip factories, implementation of any digital electronic technology on silicon wafers is essential for its commercialisation.
Remarkably, the ULTRARAM™ on silicon devices actually outperform previous incarnations of the technology on GaAs compound semiconductor wafers, demonstrating (extrapolated) data storage times of at least 1000 years, fast switching speed (for device size) and program-erase cycling endurance of at least 10 million, which is one hundred to one thousand times better than flash.
Story Source:
Materials provided by Lancaster University. Note: Content may be edited for style and length. More

The world we experience is governed by classical physics. How we move, where we are, and how fast we’re going are all determined by the classical assumption that we can only exist in one place at any one moment in time.
But in the quantum world, the behavior of individual atoms is governed by the eerie principle that a particle’s location is a probability. An atom, for instance, has a certain chance of being in one location and another chance of being at another location, at the same exact time.
When particles interact, purely as a consequence of these quantum effects, a host of odd phenomena should ensue. But observing such purely quantum mechanical behavior of interacting particles amid the overwhelming noise of the classical world is a tricky undertaking.
Now, MIT physicists have directly observed the interplay of interactions and quantum mechanics in a particular state of matter: a spinning fluid of ultracold atoms. Researchers have predicted that, in a rotating fluid, interactions will dominate and drive the particles to exhibit exotic, never-before-seen behaviors.
In a study published today in Nature, the MIT team has rapidly rotated a quantum fluid of ultracold atoms. They watched as the initially round cloud of atoms first deformed into a thin, needle-like structure. Then, at the point when classical effects should be suppressed, leaving solely interactions and quantum laws to dominate the atoms’ behavior, the needle spontaneously broke into a crystalline pattern, resembling a string of miniature, quantum tornadoes.
“This crystallization is driven purely by interactions, and tells us we’re going from the classical world to the quantum world,” says Richard Fletcher, assistant professor of physics at MIT. More

A new software tool can accelerate materials science research by cutting out tedious background research on material properties. Penn State and Sandia National Laboratories researchers recently debuted propSym, an open-source software on the programming platform MATLAB, to calculate the fundamental constants needed to describe the physical properties of solids, such as metals, ceramics or composites.
Researchers input a material’s physical characteristics and structure, and the program produces its fundamental property constants — key values researchers need to model various materials.
“Some physical models contain hundreds or thousands of redundant components, which can make the model overwhelming,” said Anubhav Roy, a doctoral student in engineering science and mechanics in the Penn State College of Engineering and first author on the paper. “The program is able to greatly reduce the number of components for any physical property that is connected to solids with inherent crystalline symmetry.”
The researchers developed propSym, the details of which were published in the Journal of Applied Crystallography, after they could not find reliable information about langasite — a material used in sensing and energy harvesting devices — in a separate joint study with Sandia National Labs.
“Traditionally, the relationships between fundamental constants and material symmetries are found only in appendices of textbooks or tables in journal articles,” said Christopher Kube, assistant professor of engineering science and mechanics at Penn State, who led the project. “After a thorough search, we were not able to find reference data for several nonlinear material properties for langasite. When data were available, we found instances of typos and inconsistencies across references. Incorrect input data will ruin a model.”
Kube and his collaborators used propSym to determine the properties of langasite, such as elasticity and the ability to accumulate electric charge. But Kube emphasized the program is not limited to those two properties alone.
“The software is adaptable to nearly any physical property of interest; the possibilities really are endless,” Kube said. “Ultimately, I hope propSymhelps to lower the entry barrier for analytical modeling of complex physical behavior. A lot of modern problems in the sciences often are deemed too challenging for analytical models without serious consideration of an analytical approach.”  
Story Source:
Materials provided by Penn State. Original written by Mariah Chuprinski. Note: Content may be edited for style and length. More

Homeowners know that the type of windows in a house contribute greatly to heating and cooling efficiency. And that’s a big deal — maintaining indoor temperatures consumes great amounts of energy and accounts for 20 to 40 percent of the national energy budgets in developed countries.
New research from the University of Pittsburgh and the University of Oxford takes energy efficient windows a step further by proposing a new “smart window” design that would harvest the sun’s energy in the winter to warm the house and reflect it in the summer to keep it cool. The work was recently published in the journal ACS Photonics and funded as part of the EPSRC Wearable and Flexible Technologies Collaboration.
“The major innovation is that these windows can change according to seasonal needs,” explained Nathan Youngblood, assistant professor of electrical and computer engineering at Pitt and first author. “They absorb near infrared light from the sun in the winter and turn it into heat for the inside of a building. In the summer months, the sun can be reflected instead of absorbed.”
The film is made up of an optical stack of materials less than 300 nanometers thick, with a very thin active layer made of “phase change” materials that can absorb the invisible wavelengths of the sun’s light and emit it as heat. That same material can be “switched” so that it turns those wavelengths of light away instead.
“Importantly, visible light is transmitted almost identically in both states, so you wouldn’t notice the change in the window,” Youngblood noted. “That aesthetic consideration is critical for the adoption of green technologies.”
The material could even be adjusted so that, for example, 30 percent of the material is turning away heat while 70 percent is absorbing and emitting it, allowing for more precise temperature control.
Harish Bhaskaran, professor at Oxford’s Materials Department, who led the research as well as the WAFT consortium said, “Here, we exploit tuning how invisible wavelengths are transmitted or reflected to modulate temperature. These ideas have come to fruition with the aid of our long-standing industrial collaborators, and are the result of long-term research.”
The researchers estimate that using these windows — including the energy required to control the film — would save 20 to 34 percent in energy usage annually compared to double-paned windows typically found in homes.
In order to create and test their prototypes, the researchers worked with Bodle Technologies, a company that specializes in ultra-thin reflective films that can function as displays by controlling color and light, as well as Eckersley O’Callaghan, a leading engineering and architectural firm, and Plasma App, a thin films company.
“This work demonstrates yet another interesting optoelectronic application of Phase Change Materials with the potential to significantly improve our everyday life,” said Peiman Hosseini, CEO of Bodle Technologies. “The commercialization of PCM-based tuneable low-e glass panels still has a number of significant challenges left to overcome; however, these preliminary results prove that the long developmental road ahead is certainly warranted. I believe this technology should be part of any future holistic policy approach tackling climate change.”
Story Source:
Materials provided by University of Pittsburgh. Original written by Maggie Lindenberg. Note: Content may be edited for style and length. More

As part of an experiment to measure — to an extremely precise degree — the charge-to-mass ratios of protons and antiprotons, the RIKEN-led BASE collaboration at CERN, Geneva, Switzerland, has found that, within the uncertainty of the experiment, matter and antimatter respond to gravity in the same way.
Matter and antimatter create some of the most interesting problems in physics today. They are essentially equivalent, except that where a particle has a positive charge its antiparticle has a negative one. In other respects they seem equivalent. However, one of the great mysteries of physics today, known as “baryon asymmetry,” is that, despite the fact that they seem equivalent, the universe seems made up entirely of matter, with very little antimatter. Naturally, scientists around the world are trying hard to find something different between the two, which could explain why we exist.
As part of this quest, scientists have explored whether matter and antimatter interact similarly with gravity, or whether antimatter would experience gravity in a different way than matter, which would violate Einstein’s weak equivalence principle. Now, the BASE collaboration has shown, within strict boundaries, that antimatter does in fact respond to gravity in the same way as matter.
The finding, published in Nature, actually came from a different experiment, which was examining the charge-to-mass ratios of protons and antiprotons, one of the other important measurements that could determine the key difference between the two.
This work involved 18 months of work at CERN’s antimatter factory. To make the measurements, the team confined antiprotons and negatively charged hydrogen ions, which they used as a proxy for protons, in a Penning trap. In this device, a particle follows a cyclical trajectory with a frequency, close to the cyclotron frequency, that scales with the trap’s magnetic-field strength and the particle’s charge-to-mass ratio. By feeding antiprotons and negatively charged hydrogen ions into the trap, one at a time, they were able to measure, under identical conditions, the cyclotron frequencies of the two particle types, comparing their charge-to-mass ratios. According to Stefan Ulmer, the leader of the project, “By doing this, we were able to obtain a result that they are essentially equivalent, to a degree four times more precise than previous measures. To this level of CPT invariance, causality and locality hold in the relativistic quantum field theories of the Standard Model.”
Interestingly, the group used the measurements to test a fundamental physics law known as the weak equivalence principle. According to this principle, different bodies in the same gravitational field should undergo the same acceleration in the absence of frictional forces. Because the BASE experiment was placed on the surface of the Earth, the proton and antiproton cyclotron-frequency measurements were made in the gravitational field on the Earth’s surface, and any difference between the gravitational interaction of protons and antiprotons would result in a difference between the cyclotron frequencies.
By sampling the gravitational field of the Earth as the planet orbited the Sun, the scientists found that matter and antimatter responded to gravity in the same way up to a degree of three parts in 100, which means that the gravitational acceleration of matter and antimatter are identical within 97% of the experienced acceleration.
Ulmer adds that these measurements could lead to new physics. He says, “The 3% accuracy of the gravitational interaction obtained in this study is comparable to the accuracy goal of the gravitational interaction between antimatter and matter that other research groups plan to measure using free-falling anti-hydrogen atoms. If the results of our study differ from those of the other groups, it could lead to the dawn of a completely new physics.”
The research group, led by RIKEN, included scientists from international partners including CERN, the Max Planck Society, the National Metrology Institute in Germany PTB, the Universities of Mainz and Hannover, the University of Tokyo, and GSI Darmstadt.
Story Source:
Materials provided by RIKEN. Note: Content may be edited for style and length. More

Topological materials move electrons along their surface and edges without any loss, making them promising materials for dissipationless, high-efficiency electronics. Researchers are especially interested in using these materials as transistors, the backbone of all modern electronics. But there’s a problem: Transistors switch electronic current on and off, but it’s difficult to turn off the dissipationless flow of electrons in topological materials.
Now, Harvard University researchers have designed and simulated the first topological acoustic transistors — with sound waves instead of electrons — and proposed a connection architecture to form a universal logic gate that can switch the flow of sound on and off.
“Since the advent of topological materials around 2007, there has been a lot of interest in developing a topological electronic transistor,” said Jenny Hoffman, the Clowes Professor of Science at the Harvard John A. Paulson School of Engineering and Applied Sciences (SEAS) and the Department of Physics. “Although the materials we used won’t yield an electronic topological transistor, our general design process applies to both quantum materials and photonic crystals, raising hopes that electronic and optical equivalents may not be far behind.”
The research is published in Physical Review Letters.
By using acoustic topological insulators, the researchers were able to sidestep the complicated quantum mechanics of electron topological insulators.
“The equations for sound waves are exactly solvable, which allowed us to numerically find just the right combination of materials to design a topological acoustic waveguide that turns on when heated, and off when cooled,” said Harris Pirie, a former graduate student in the Department of Physics and first author of the paper. More