Toggle light / dark theme

Get the latest international news and world events from around the world.

Log in for authorized contributors

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.

Rise of the robots: The promise of physical AI

A pair of swiveling, human-like robotic arms, built for physical artificial intelligence research, mirror the motions of an operator in a VR headset twirling his hands like a magician.

With enough practice, arms like these can complete everyday tasks alone, says Tokyo company Enactic, which is developing humanoid robots to wash dishes and do laundry in short-staffed Japanese .

Welcome to the future of AI as it starts to infiltrate the material world in the form of smart robots, self-driving cars and other autonomous machines.

Traces of bacteria inside brain tumors may affect tumor behavior

Researchers at The University of Texas MD Anderson Cancer Center have uncovered unexpected traces of bacteria within brain tumors. This discovery offers new insights into the environment in which brain tumors grow and sets the stage for future studies seeking to improve treatment outcomes.

Published today in Nature Medicine, the data revealed that bacterial genetic and cellular elements were present inside brain tumor cells and across the tumor microenvironment. These bacterial components appeared biologically active, potentially influencing tumor behavior and progression in patients with gliomas and brain metastases.

The multi-institutional study was led by Golnaz Morad, D.D.S, Ph.D., postdoctoral research fellow in Surgical Oncology, and Jennifer Wargo, M.D., professor of Surgical Oncology and Genomic Medicine and core member of the James P. Allison Institute—working in close collaboration with MD Anderson’s Platform for Innovative Microbiome and Translational Research (PRIME-TR).

How the human brain anticipates and regulates the body’s needs

They also used a recently validated map of deep brain areas. This in vivo atlas, Brainstem Navigator, maps the regions involved in regulating the autonomic, immune and endocrine systems.

The authors analytic approach was guided by decades of basic research that has identified two main brain pathways in mammals: one set of pathways (allostatic) that sends signals from the brain to control the body’s organs, and the other set (interoceptive) that sends signals from the body to the brain, informing it about what’s happening inside us.

The findings replicated and expanded on their previous 3 Tesla work, confirming nearly all the direct connections identified in non-human mammals: 100% of those between cortical areas and 96% of those linking subcortical areas to both cortical and other subcortical areas. As expected, the authors found two-way connections between the brain areas that help manage the body’s needs (like the anterior cingulate cortex) and the areas that sense what’s happening inside the body (like the posterior insula). This means these regions communicate back and forth, helping the brain predict and regulate what the body needs.

Mounting evidence suggests that one of the brain’s central roles is to anticipate and meet the body’s energy needs. The findings place the monitoring and regulation of the body’s needs at the functional core of the human brain, showing the close connection between mental and physical health.


Previous studies in both animal models and humans have pointed to the existence of a distributed system in the brain that helps it anticipate and prepare for the body’s energy needs — a process called allostasis — as well as monitor the sensory conditions inside the body, known as interoception.

In an earlier study using 3 Tesla fMRI, the team mapped a network supporting allostasis and interoception in the human brain, but the comparatively limited spatial resolution and sensitivity of the 3 Tesla technology made it difficult to fully capture the system’s smaller structures in the brainstem, which are known to play a key role in these processes.

AI-powered LED system delivers stable wireless power for indoor IoT devices

The world’s first automatic and adaptive, dual-mode light-emitting diode (LED)-based optical wireless power transmission system, that operates seamlessly under both dark and bright lighting conditions, has been developed by scientists at Science Tokyo. The system, along with artificial intelligence-powered image recognition, can efficiently power multiple devices in order without interruption. Because it is LED-based, it offers a low-cost and safe solution ideal for building sustainable indoor Internet of Things infrastructure.

With the rapid development of Internet of Things (IoT), the demand for efficient and flexible power solutions is also increasing. Traditional power delivery methods, such as batteries and cable connections, have many drawbacks. Batteries need frequent charging and replacement, while cables restrict device mobility.

Optical wireless power transmission (OWPT) is an emerging technology that can address these limitations. In OWPT, energy is transmitted through , without physical wires, by converting electricity to light, transmitting it, and then reconverting light back into using photovoltaic (PV) receivers.

Chang’e-6 samples reveal first evidence of impact-formed hematite and maghemite on the moon

A joint research team from the Institute of Geochemistry of the Chinese Academy of Sciences (IGCAS) and Shandong University has for the first time identified crystalline hematite (α-Fe2O3) and maghemite (γ-Fe2O3) formed by a major impact event in lunar soil samples retrieved by China’s Chang’e-6 mission from the South Pole–Aitken (SPA) Basin. This finding, published in Science Advances on November 14, provides direct sample-based evidence of highly oxidized materials on the lunar surface.

Redox reactions are a fundamental component of planetary formation and evolution. Nevertheless, scientific studies have shown that neither the oxygen fugacity of the lunar interior nor the environment favors oxidation. Consistent with this, multivalent iron on the moon primarily exists in its ferrous (Fe2+) and metallic (Fe0) states, suggesting an overall reduced state. However, with further lunar exploration, recent orbital remote sensing studies using visible-near-infrared spectroscopy have suggested the widespread presence of hematite in the moon’s high-latitude regions.

Furthermore, earlier research on Chang’e-5 samples first revealed impact-generated sub-micron magnetite (Fe3O4) and evidence of Fe3+ in impact glasses. These results indicate that localized oxidizing environments on the moon existed during lunar surface modification processes driven by external impacts.

AI math genius delivers 100% accurate results

At the 2024 International Mathematical Olympiad (IMO), one competitor did so well that it would have been awarded the Silver Prize, except for one thing: it was an AI system. This was the first time AI had achieved a medal-level performance in the competition’s history. In a paper published in the journal Nature, researchers detail the technology behind this remarkable achievement.

The AI is AlphaProof, a sophisticated program developed by Google DeepMind that learns to solve complex mathematical problems. The achievement at the IMO was impressive enough, but what really makes AlphaProof special is its ability to find and correct errors. While (LLMs) can solve , they often can’t guarantee the accuracy of their solutions. There may be hidden flaws in their reasoning.

AlphaProof is different because its answers are always 100% correct. That’s because it uses a specialized software environment called Lean (originally developed by Microsoft Research) that acts like a strict teacher verifying every logical step. This means the computer itself verifies answers, so its conclusions are trustworthy.

/* */