A resilience-focused deep reinforcement learning strategy reduced cumulative network performance loss by up to 79.7% relative to conventional signal control in benchmark tests and required up to 45.7% fewer behavioral-adjustment iterations to reach a new equilibrium under severe disruption.

- KAIST and MIT Develop the World’s First Transmissive Mid-Infrared Spatial Light Modulator with Independently Controllable Pixels

- Multiple Sensor Functions on a Single Optical Chip... Laying the Foundation for the Era of “Software-Defined Sensors” and Expected to Bring Innovation to Sensors for Satellites, Launch Vehicles, and Space Stations

- A research team led by Distinguished Professor Sang Yup Lee from the Department of Chemical and Biomolecular Engineering analyzed the key bottlenecks limiting the commercialization of biomanufacturing, and proposed an AI-based roadmap for industrialization

- Analyzed case studies of succinic acid and biodegradable plastic (PHA), proposing solutions to bridge the "valley of death" between laboratory and industry

Arsenic contamination in paddy soils threatens food safety because flooded conditions can mobilize the toxic element and make it available for uptake by rice. A new study suggests that biochar coated with chitosan, a natural material derived from chitin, could help trap arsenic in soil and prevent it from reaching the edible grain.

Biochar is widely promoted as a tool for improving soil health and storing carbon, but its effectiveness can vary greatly from one field to another. A new global study reveals that the amount of nitrogen already present in soil is a key factor controlling both how much carbon biochar stores and how that carbon becomes stabilized.

Biochar is widely promoted as a tool for improving soil health and storing carbon. Yet its effects on soil carbon are not always consistent. New research published in Biochar shows that part of the explanation may lie in highly reactive molecules produced by persistent free radicals within biochar.

(IDAHO FALLS, Idaho) – For the first time, researchers from Idaho National Laboratory, the Grainger College of Engineering at the University of Illinois Urbana-Champaign, and Purdue University used a geographically distributed control system to remotely and automatically adjust the power of a research reactor in real time, with the reactor’s own safety systems retaining full control throughout. Working with PUR1, a low-power research reactor at Purdue, the team first demonstrated remote power adjustments through an automatic adjustment system called a digital control loop, then made the loop further autonomous using a reinforcement learning model running in software that simulates how physical forces interact within the reactor. The demonstration linked three sites operating together in real time: high-performance computing systems in Idaho, the PUR1 reactor in Indiana, and a Microsoft Azure cloud environment in Virginia.

Virginia Tech researchers have developed an artificial intelligence (AI) model designed to help policymakers manage growing competition between agriculture and semiconductor manufacturing, as drough strains water supplies across the United States.