Researchers trained an AI to assess post-disaster building damage just by looking at aerial images of the aftermath.
The researchers at USC have made some discoveries about the network behind the Panama Papers, uncovering uniquely fragmented network behavior and transactions. This is vastly different from more traditional social or organizational networks, demonstrating why these systems of transactions and associations are so robust and difficult to infiltrate or take down
Researchers at Osaka City University create a quantum algorithm that removes spin contaminants while making chemical calculations on quantum computers. This allows for predictions of electronic and molecular behavior with degrees of precision not achievable with classical computers and paves the way for practical quantum computers to become a reality.
Carbon-based computers have the potential to be a lot faster and much more energy efficient than silicon-based computers, but 2D graphene and carbon nanotubes have proved challenging to turn into the elements needed to construct transistor circuits. Graphene nanoribbons can overcome these limitations, but to date scientists have been made only semiconductors and insulators, not the metallic wires to connect them. UC Berkeley scientists have now achieved the goal of a metallic graphene nanoribbon.
Engineers at CSEM have developed a new machine-learning method that paves the way for artificial intelligence to be used in applications that until now have been deemed too sensitive. The method, which has been tested by running simulations on a climate-control system for a 100-room building, is poised to deliver energy savings of around 20%.
The Greyhound framework, named after the breed of dogs known for their hunting abilities, was designed and implemented by an SUTD-led research team to systematically sniff out security lapses in Wi-Fi and Bluetooth enabled devices.
The turbulence code GENE (Gyrokinetic Electromagnetic Numerical Experiment), developed at Max Planck Institute for Plasma Physics (IPP) at Garching, Germany, has proven to be very useful for the theoretical description of turbulence in the plasma of tokamak-type fusion devices. Extended for the more complex geometry of stellarator-type devices, computer simulations with GENE now indicate a new method to reduce plasma turbulence in stellarator plasmas. This could significantly increase the efficiency of a future fusion power plant.
Scientists at Shenzhen University have recently developed an all-optical ultrafast imaging system with high spatial and temporal resolutions, as well as a high frame rate. Because the method is all-optical, it's free from the bottlenecks that arise from scanning with mechanical and electronic components.
Artificial intelligence researchers have improved the performance of deep neural networks by combining feature normalization and feature attention modules into a single module that they call attentive normalization. The hybrid module improves the accuracy of the system significantly, while using negligible extra computational power.
Researchers have discovered that a new material can act as a super-fast magnetic switch. When struck by successive ultra-short laser pulses it exhibits 'toggle switching' that could increase the capacity of the global fibre optic cable network by an order of magnitude.