Computer modeling

MIT researchers have developed a new technique that uses a large language model to allow robots to self-correct after making a mistake, reports Brian Heater for TechCrunch. “Researchers behind the study note that while imitation learning (learning to do a task through observation) is popular in the world of home robotics, it often can’t account for the countless small environmental variations that can interfere with regular operation, thus requiring a system to restart from square one,” writes Heater. “The new research addresses this, in part, by breaking demonstrations into smaller subsets." 

MIT researchers are hoping to use Dyson maps “to translate the language of classical physics into terms that a quantum computer—a machine designed to solve complex quandaries by leveraging the unique properties of quantum particles—can understand,” reports Darren Orf for Popular Mechanics. 

MIT researchers have developed SoftZoo, “an open framework platform that simulated a variety of 3D model animals performing specific tasks in multiple environmental settings,” reports Andrew Paul for Popular Science. “This computational approach to co-designing the soft robot bodies and their brains (that is, their controllers) opens the door to rapidly creating customized machines that are designed for a specific task,” says CSAIL director, Prof. Daniela Rus.

Researchers at MIT have developed “SoftZoo,” a platform designed to “study the physics, look and locomotion and other aspects of different soft robot models,” reports Brian Heater for TechCrunch. “Dragonflies can perform very agile maneuvers that other flying creatures cannot complete because they have special structures on their wings that change their center of mass when they fly,” says graduate student Tsun-Hsuan Wang. “Our platform optimizes locomotion the same way a dragonfly is naturally more adept at working through its surroundings.”

National Geographic reporter Maya Wei-Haas explores how the ancient art of origami is being applied to fields such a robotics, medicine and space exploration. Wei-Haas notes that Prof. Daniela Rus and her team developed a robot that can fold to fit inside a pill capsule, while Prof. Erik Demaine has designed complex, curving fold patterns. “You get these really impressive 3D forms with very simple creasing,” says Demaine.

MIT researchers have created Thesan, the most detailed model of the early universe to date, reports New Scientist. “Thesan shows how radiation shaped the universe from 400,000 to 1 billion years after the Big Bang,” writes New Scientist.

MIT researchers have developed a new computational model that could be used to help explain differences in how neurotypical adults and adults with autism recognize emotions via facial expressions, reports Tony Ho Tran for The Daily Beast. “For visual behaviors, the study suggests that [the IT cortex] pays a strong role,” says research scientist Kohitij Kar. “But it might not be the only region. Other regions like amygdala have been implicated strongly as well. But these studies illustrate how having good [AI models] of the brain will be key to identifying those regions as well.”

Scientists from MIT and other institutions have developed the largest, most detailed computer model of the universe’s first billion years, which could help shed light on how the early universe evolved, reports Charles Q. Choi for Scientific American. The model, named THESAN, “can track the birth and evolution of hundreds of thousands of galaxies within a cubic volume spanning more than 300 million light-years across.”

TechCrunch reporter Brian Heater spotlights new MIT robotics research, including a team of CSAIL researchers “working on a system that utilizes a robotic arm to help people get dressed.” Heater notes that the “issue is one of robotic vision — specifically finding a method to give the system a better view of the human arm it’s working to dress.”

Researchers from MIT and other institutions have developed a new simulation that illuminates how stars formed in the early universe, reports Martin Finucane for The Boston Globe. “It was a neutral, dark cosmos that became bright and ionized as light began to emerge from the first galaxies,” explains Aaron Smith, a NASA Einstein Fellow in MIT’s Kavli Institute for Astrophysics and Space Research.

MIT researchers have developed a new simulation of the early universe, shedding light onto the period when the first stars were formed, reports Audrey Carleton for Vice. “Using existing models of the early universe and of cosmic dust, matched with new code created to interpret how light and gas interacted with one another, they created a visual depiction of the growth of the universe,” writes Carleton.

Axios reporter Alison Snyder writes that a new study by MIT researchers demonstrates how AI algorithms could provide insight into the human brain’s processing abilities. The researchers found “Predicting the next word someone might say — like AI algorithms now do when you search the internet or text a friend — may be a key part of the human brain's ability to process language,” writes Snyder.

Using an integrative modeling technique, MIT researchers compared dozens of machine learning algorithms to brain scans as part of an effort to better understand how the brain processes language. The researchers found that “neural networks and computational science might, in fact, be critical tools in providing insight into the great mystery of how the brain processes information of all kinds,” writes Anna Blaustein for Scientific American.

Washington Post reporter Paige Winfield Cunningham spotlights how MIT researchers have developed a new way to estimate the impact of Covid-19. The researchers “developed a way to compare and merge more than two dozen different models from universities and analytics groups around the country.”

Research scientist Brian Subirana speaks with The Economist’s Babbage podcast about his work developing a new AI system that could be used to help diagnose people asymptomatic Covid-19.