Toggle light / dark theme

Sustainable AI: Physical neural networks exploit light to train more efficiently

Artificial intelligence is now part of our daily lives, with the subsequent pressing need for larger, more complex models. However, the demand for ever-increasing power and computing capacity is rising faster than the performance traditional computers can provide.

To overcome these limitations, research is moving towards innovative technologies such as physical neural networks, analog circuits that directly exploit the laws of physics (properties of light beams, quantum phenomena) to process information. Their potential is at the heart of the study published in the journal Nature. It is the outcome of collaboration between several international institutes, including the Politecnico di Milano, the École Polytechnique Fédérale in Lausanne, Stanford University, the University of Cambridge, and the Max Planck Institute.

The article entitled “Training of Physical Neural Networks” discusses the steps of research on training physical neural networks, carried out with the collaboration of Francesco Morichetti, professor at DEIB—Department of Electronics, Information and Bioengineering, and head of the university’s Photonic Devices Lab.

Quantum calculations provide a sharper image of subatomic stress

Stress is a very real factor in the structure of our universe. Not the kind of stress that students experience when taking a test, but rather the physical stresses that affect everyday objects. Consider the stress that heavy vehicles exert on a bridge as they cross over it—it’s essential that engineers understand and consider this factor when designing new trestles. Or consider the stresses that a star experiences—this internal factor influences everything from its shine to its lifetime.

Something From Nothing — Physicists Mimic the “Impossible” Schwinger Effect

Superfluid helium reveals a manageable analog to the Schwinger effect. It deepens understanding of vortices and quantum tunneling. In 1951, physicist Julian Schwinger proposed that applying a constant electric field to a vacuum could cause electron-positron pairs to emerge spontaneously, a proces

Physicists Measured The Pulse of an Atom’s Magnetic Heart in Real Time

The pulse of an atom’s magnetic heart as it ticks back and forth between quantum states has been timed in a laboratory.

Physicists used a scanning tunneling microscope to observe electrons as they moved in sync with the nucleus of an atom of titanium-49, allowing them to estimate the duration of the core’s magnetic beat in isolation.

“These findings,” they write in their paper, “give an atomic-scale insight into the nature of nuclear spin relaxation and are relevant for the development of atomically assembled qubit platforms.”

Non-gaussian States Of Light Unlock Universal Computation With Enhanced Success Probabilities And Optimised Photon Requirements

Non-Gaussian states of light represent a crucial component for advancements in quantum technologies, holding immense potential for universal computation, robust error correction, and highly sensitive sensing, yet creating these states remains a significant challenge. Fumiya Hanamura, Kan Takase, and Hironari Nagayoshi, along with their colleagues, now present a new approach to overcome these hurdles, introducing ‘non-Gaussian control parameters’ that offer a more effective way to measure and optimise the generation of these complex states. This method moves beyond traditional benchmarks, such as stellar rank, by providing a continuous and practical measure of non-Gaussianity, and importantly, dramatically reduces the resources needed for successful state creation. Demonstrations across a range of states, including cat states and GKP states, reveal that this technique cuts required photon detections by a factor of three and boosts preparation probability, paving the way for more feasible and scalable quantum technologies and fault-tolerant computation.


Researchers have developed a new method for generating complex states of light that significantly reduces the resources needed for advanced technologies like quantum computing and sensing, achieving a threefold reduction in required measurements and a substantial increase in success rates across various light states.

Michio Kaku: This could finally solve Einstein’s unfinished equation | Full Interview

“An equation, perhaps no more than one inch long, that would allow us to, quote, ‘Read the mind of God.’”

Up next, Michio Kaku: The Universe in a Nutshell (Full Presentation) ► • Michio Kaku: The Universe in a Nutshell (F…

What if everything we know about computing is on the verge of collapsing? Physicist Michio Kaku explores the next wave that could render traditional tech obsolete: Quantum computing.

Quantum computers, Kaku argues, could unlock the secrets of life itself: and could allow us to finally advance Albert Einstein’s quest for a theory of everything.

00:00:00 Quantum computing and Michio’s book Quantum Supremacy00:01:19 Einstein’s unfinished theory.
00:03:45 String theory as the \.

A new way to control terahertz light for faster electronics

In a breakthrough for next-generation technologies, scientists have learned how to precisely control the behavior of tiny waves of light and electrons, paving the way for faster communications and quantum devices.

Controlling light at the smallest scales is crucial for creating incredibly small, fast and efficient devices. Instead of bulky wires and circuits, we can use light to transmit information. One challenge of this approach is that light, with its relatively large wavelength, is not easily confined to small spaces.

However, in a study published in the journal Light: Science & Applications, researchers have developed a method to control tiny waves of light and electrons called Dirac plasmon polaritons (DPPs).

Quantum dot and polymer cross-linking enables 50% stretch capability for micro-LED displays

A research team has developed a next-generation display core material with excellent stretchability and superior color reproduction. The team developed a high-performance color-conversion layer that is more flexible and vivid than conventional ones. This layer was successfully applied to the development of a stretchable micro-LED display, drawing significant attention.

The paper is published in the journal Advanced Materials. The team was led by Professor Jiwoong Yang in the Department of Energy Science and Engineering at DGIST and included Professor Moonkee Choi and Professor Jongnam Park of UNIST and Professor Daehyeong Kim of Seoul National University.

Professor Yang’s team has recently developed, for the first time in the world, a new technology that enables the direct linkage of quantum dots, which are emerging as next-generation materials, with stretchable polymers that can stretch like rubber.

Core technology developed for ultra-high-resolution quantum dot displays

A research team has developed a direct optical lithography (DOL) technology that patterns quantum dots (QDs) at ultra-high resolution using only light, without photoresist. Through this, they also provided guidelines for selecting cross-linkers essential for fabricating high-performance QLEDs. This achievement is regarded as a core fundamental technology that can be applied to a wide range of optoelectronic devices, including micro-QLEDs, ultra-high-resolution displays, transparent electronic devices, and next-generation image sensors.

From layered transition metal oxide to 2D material: Scientists make 2H-NbO₂ discovery

2H-NbO₂—a novel van der Waals oxide synthesized by researchers from Japan—exhibits strongly correlated electronic properties with two-dimensional flexibility. By chemically extracting lithium ions from the layered sheets of LiNbO₂, the researchers transformed a three-dimensional oxide into a two-dimensional material—unlocking unique properties like Mott insulating states and superconductivity. Bridging transition metal oxides and 2D materials, the discovery paves the way for realizing advanced quantum materials in next-generation electronic devices.

Two-dimensional (2D) materials have become a cornerstone of next-generation electronic research. These materials—with their layers held together by weak van der Waals (vdW) forces—are celebrated for their unique quantum properties and promising applications in electronics. However, despite significant progress in 2D materials like graphene and , one attractive family of materials called “” or TMOs, remains unexplored for 2D application.

TMOs are a versatile class of materials known for their complex like superconductivity, magnetism, and metal-insulator transitions. But due to their inherently strong ionic bonding, these oxides do not typically form vdW structures and therefore remain absent from 2D materials basically.

/* */