Lifeboat Foundation

Neural radiance fields (NeRFs) are advanced machine learning techniques that can generate three-dimensional (3D) representations of objects or environments from two-dimensional (2D) images. As these techniques can model complex real-world environments realistically and in detail, they could greatly support robotics research.

Most existing datasets and platforms for training NeRFs, however, are designed to be used offline, as they require the completion of a pose optimization step that significantly delays the creation of photo realistic representations. This has so far prevented most roboticists from using these techniques to test their algorithms on physical robots in real-time.

Continue reading “A software package to ease the use of neural radiance fields in robotics research” »

Identifying new sources that produce electrons faster could help to advance the many imaging techniques that rely on electrons. In a recent paper published in Physical Review Letters, a team of researchers at Eindhoven University of Technology demonstrated the scattering of subpicosecond electron bunches from an ultracold electron source.

“Our research group is working to develop the next generation of ultrafast electron sources to push imaging techniques such as ultrafast electron diffraction to the next level,” Tim de Raadt, one of the researchers who carried out the study, told Phys.org.

“The idea of using laser-cooled ultracold gas clouds as an electron source to improve the state-of-the-art in brightness was first introduced in a paper published in 2005. Since then, research efforts have produced multiple versions of such a ultracold electron source, with the most recent one (used in this work) focusing on making the source compact, easy to align and operate, and being more stable, as described in another past paper that also studied the transverse electron beam properties.”

The first protein-based nano-computing agent that functions as a circuit has been created by Penn State researchers. The milestone puts them one step closer to developing next-generation cell-based therapies to treat diseases like diabetes and cancer.

Traditional synthetic biology approaches for cell-based therapies, such as ones that destroy cancer cells or encourage tissue regeneration after injury, rely on the expression or suppression of proteins that produce a desired action within a cell. This approach can take time (for proteins to be expressed and degrade) and cost cellular energy in the process. A team of Penn State College of Medicine and Huck Institutes of the Life Sciences researchers are taking a different approach.

“We’re engineering proteins that directly produce a desired action,” said Nikolay Dokholyan, G. Thomas Passananti Professor and vice chair for research in the Department of Pharmacology. “Our protein-based devices or nano-computing agents respond directly to stimuli (inputs) and then produce a desired action (outputs).”

While hearing aids and cochlear implants offer limited relief, no available treatment can reverse or prevent this group of genetic conditions, prompting scientists to evaluate gene therapies for alternative solutions.

One of the most promising tools used in these therapies—adeno associated virus (AAV) vectors—has galvanized the hearing-loss community in recent years.

Excitations in solids can also be represented mathematically as quasiparticles; for example, lattice vibrations that increase with temperature can be well described as phonons. Mathematically, also quasiparticles can be described that have never been observed in a material before. If such “theoretical” quasiparticles have interesting talents, then it is worth taking a closer look. Take fractons, for example.

Fractons are fractions of spin excitations and are not allowed to possess kinetic energy. As a consequence, they are completely stationary and immobile. This makes fractons new candidates for perfectly secure information storage. Especially since they can be moved under special conditions, namely piggyback on another quasiparticle.

“Fractons have emerged from a mathematical extension of quantum electrodynamics, in which electric fields are treated not as vectors but as tensors—completely detached from real materials,” explains Prof. Dr. Johannes Reuther, theoretical physicist at the Freie Universität Berlin and at HZB.

Researchers move an individual Mg+ ion more than 100,000 times between different sites in a trapping array without dropping it or ruining its quantum coherence.

A combination of two techniques provides warning signs that the stress on a material will lead to failure.

Soft elastomers, such as rubber, plastic, and silicone, are used in thousands of products, such as gaskets, hoses, and inflatable rafts, but under stress, these materials tend to crack abruptly, without warning. Now, using an improved method to image structural changes in a sample under stress, researchers have shown that a subtle pattern of molecular motions at the surface of the material occurs several minutes before a final failure [1]. With development, they believe the technique may help engineers monitor materials while in use and detect failures well before they happen. The researchers also showed that their approach works for some more brittle polymer materials.

When studying the mechanical failure of a material, researchers often experiment by cutting a small notch into a thin sheet of the material and applying a slowly increasing force that pulls the notch apart. Eventually, a crack will grow and spread rapidly from the notch. Materials scientist Costantino Creton of Paris Sciences and Letters University says that over the past few years, such experiments have led to two general findings for elastomers. First, by embedding light-emitting, force-sensitive molecules into test materials, researchers have shown that, prior to crack initiation, irreversible molecular-bond damage accumulates very close to the initial notch (within 0.1 mm). Second, using sensitive spectroscopy techniques, other studies have found signs of unusual microscopic rearrangements of the polymer molecules occurring over larger regions of the material just prior to failure.

Single electrons stay stationary in superconductors with “flat-band” electronic structures, which could lead to low-energy-consumption devices made from such materials.

In 2018, researchers discovered that two layers of graphene, stacked and twisted at a specific angle, could exhibit superconductivity. Theorists have determined that the electronic structure of such a twisted material approximately resembles a “flat band,” which means that the energy of the materials’ free electrons remains constant regardless of the electrons’ momenta. This phenomenon inspired a flurry of work on systems that exhibit flat-band superconductivity. However, most of the research has focused on how such systems behave under equilibrium conditions. Now Päivi Törmä of Aalto University in Finland and her colleagues have probed the behavior of superconducting flat-band systems under nonequilibrium conditions [1]. The findings could help in the design of superconducting devices with low energy consumption.

Törmä and her colleagues considered an idealized flat-band material subjected to an applied voltage, making it a nonequilibrium system. Their predictions indicate that in this nonequilibrium system the paired and unpaired electrons follow the same behavior patterns as those in an equilibrium system: unpaired electrons form stationary quasiparticles and paired electrons flow with zero resistance. Additionally, in both types of systems the flat band helps the electrons form the bound pairs required for superconductivity.

ChatGPT for iPhone launched in the US last week, with OpenAI promising that it would come to more countries “in the coming weeks.”

The next phase in the rollout has now happened earlier than expected, with 11 more countries added on Wednesday, and a further 35 today …

While you could of course access the ChatGPT website on your iPhone, an iPhone app makes it more convenient – especially as the app is free, and has no ads.