Lifeboat Foundation: Safeguarding Humanity
Toggle light / dark theme

Blog

Category: information science

Machine learning algorithm rapidly reconstructs 3D images from X-ray data

Soon, researchers may be able to create movies of their favorite protein or virus better and faster than ever before. Researchers at the Department of Energy’s SLAC National Accelerator Laboratory have pioneered a new machine learning method—called X-RAI (X-Ray single particle imaging with Amortized Inference)—that can “look” at millions of X-ray laser-generated images and create a three-dimensional reconstruction of the target particle. The team recently reported their findings in Nature Communications.

X-RAI’s ability to sort through a massive number of images and learn as it goes could unlock limits in data-gathering, allowing researchers to see molecules up close—and perhaps even on the move. “There is really no limit” to the dataset size it can handle, said SLAC staff scientist Frédéric Poitevin, one of the study’s principal investigators.

How small can optical computers get? Scaling laws reveal new strategies

The research, published in Nature Communications, addresses one of the key challenges to engineering computers that run on light instead of electricity: making those devices small enough to be practical. Just as algorithms on digital computers require time and memory to run, light-based systems also require resources to operate, including sufficient physical space for light waves to propagate, interact and perform analog computation.

Lead authors Francesco Monticone, associate professor of electrical and computer engineering, and Yandong Li, Ph.D. ‘23, postdoctoral researcher, revealed scaling laws for free-space optics and photonic circuits by analyzing how their size must grow as the tasks they perform become more complex.

Explainable AI and turbulence: A fresh look at an unsolved physics problem

While atmospheric turbulence is a familiar culprit of rough flights, the chaotic movement of turbulent flows remains an unsolved problem in physics. To gain insight into the system, a team of researchers used explainable AI to pinpoint the most important regions in a turbulent flow, according to a Nature Communications study led by the University of Michigan and the Universitat Politècnica de València.

A clearer understanding of turbulence could improve forecasting, helping pilots navigate around turbulent areas to avoid passenger injuries or structural damage. It can also help engineers manipulate turbulence, dialing it up to help industrial mixing like water treatment or dialing it down to improve fuel efficiency in vehicles.

“For more than a century, turbulence research has struggled with equations too complex to solve, experiments too difficult to perform, and computers too weak to simulate reality. Artificial Intelligence has now given us a new tool to confront this challenge, leading to a breakthrough with profound practical implications,” said Sergio Hoyas, a professor of aerospace engineering at the Universitat Politècnica de València and co-author of the study.

Quantum computers could be powerful enough to decrypt Bitcoin sometime after 2030, CEO of Nvidia’s quantum partner says

“You should have a few good years ahead of you but I wouldn’t hold my Bitcoin,” Peronnin said, laughing. “They need to fork [move to a stronger blockchain] by 2030, basically. Quantum computers will be ready to be a threat a bit later than that,” he said.

Quantum doesn’t just threaten Bitcoin, of course, but all banking encryption. And it is likely that in all these cases companies are developing quantum resistant tools to upgrade their existing security systems.

Defensive security algorithms are improving, Peronnin said, so it’s not certain when the blockchain will become vulnerable to a quantum attack. But “the threshold for such an event is coming closer to us year by year,” he said.

First full simulation of 50 qubit universal quantum computer achieved

A research team at the Jülich Supercomputing Center, together with experts from NVIDIA, has set a new record in quantum simulation: for the first time, a universal quantum computer with 50 qubits has been fully simulated—a feat achieved on Europe’s first exascale supercomputer, JUPITER, inaugurated at Forschungszentrum Jülich in September.

The result surpasses the previous world record of 48 qubits, established by Jülich researchers in 2022 on Japan’s K computer. It showcases the immense computational power of JUPITER and opens new horizons for developing and testing quantum algorithms. The research is published on the arXiv preprint server.

Quantum computer simulations are vital for developing future quantum systems. They allow researchers to verify experimental results and test new algorithms long before powerful quantum machines become reality. Among these are the Variational Quantum Eigensolver (VQE), which can model molecules and materials, and the Quantum Approximate Optimization Algorithm (QAOA), used for optimization problems in logistics, finance, and artificial intelligence.

Sharper MRI scans may be on horizon thanks to new physics-based model

Researchers at Rice University and Oak Ridge National Laboratory have unveiled a physics-based model of magnetic resonance relaxation that bridges molecular-scale dynamics with macroscopic magnetic resonance imaging (MRI) signals, promising new insight into how contrast agents interact with water molecules. This advancement paves the way for sharper medical imaging and safer diagnostics using MRI.

The study is published in The Journal of Chemical Physics.

This new approach, known as the NMR eigenmodes framework, solves the full physical equations that can be used to interpret how water molecules relax around metal-based imaging agents, a task that previous models approximated. These findings could alter the development and application of new contrast agents in both medicine and materials science.

New Quantum Algorithm Could Explain Why Matter Exists at All

Researchers used IBM’s quantum computers to create scalable quantum circuits that simulate matter under extreme conditions, offering new insight into fundamental forces and the origins of the universe. Simulating how matter behaves under extreme conditions is essential for exploring some of the d

From Data to Physics: An Agentic Large Language Model Solves a Competitive Adsorption Puzzle

We show that an agentic large language model (LLM) (OpenAI o3 with deep research) can autonomously reason, write code, and iteratively refine hypotheses to derive a physically interpretable equation for competitive adsorption on metal-organic layers (MOLs)—an open problem our lab had struggled with for months. In a single 29-min session, o3 formulated the governing equations, generated fitting scripts, diagnosed shortcomings, and produced a compact three-parameter model that quantitatively matches experiments across a dozen carboxylic acids.

Cracking the code of complexity in computer science’s P vs. NP problem

New research from the University of Waterloo is making inroads on one of the biggest problems in theoretical computer science. But the way to do it, according to Cameron Seth, a Ph.D. researcher working in the field of algorithmic approximation, is by breaking the problem down into smaller pieces.

“Everyone working in computer science and mathematics knows about the ‘P vs. NP’ problem,” Seth says. “It’s one of the notorious Millennium Prize Problems: so famous and so difficult that solving one will earn you a million dollars.”

To understand the crux of the “P vs. NP” problem, imagine an enormous jigsaw puzzle or a Sudoku puzzle. It would be a “P” problem if it could be solved relatively quickly by a computer, whereas they would be an “NP” problem if they were extremely difficult to solve, but a provided solution could be quickly verified.

Black hole mergers could give rise to observable gravitational-wave tails

Black holes, regions of spacetime in which gravity is so strong that nothing can escape, are intriguing and extensively studied cosmological phenomena. Einstein’s general theory of relativity predicts that when two black holes merge, they emit ripples in spacetime known as gravitational waves.

Once the gravitational waves originating from black hole mergers fade, subtle hints of these waves could remain, known as late-time gravitational-wave tails. While the existence of these tails has been widely theorized about in the past, it was not yet conclusively confirmed.

Researchers at Niels Bohr Institute, University of Lisbon and other institutes worldwide recently performed black hole merger simulations based on Einstein’s equations, to further probe the existence of late-time gravitational-wave tails. Their simulations, outlined in a paper in Physical Review Letters, suggest that these tails not only exist, but could also have a larger amplitude than originally predicted and could thus be observed in future experiments.

/* */