Putting clear bounds on uncertainty
Computer scientists want to know the exact limits in our ability to clean up, and reconstruct, partly blurred images.
MIT researchers develop an AI model that can detect future lung cancer risk
Deep-learning model takes a personalized approach to assessing each patient’s risk of lung cancer based on CT scans.
How Huntington’s disease affects different neurons
A new study identifies cells that are the most vulnerable within a brain structure involved in mood and movement.
How to push, wiggle, or drill an object through granular material
A method for quickly predicting the forces needed to push objects through "flowable media" could help engineers drive robots or anchor ships.
New technologies reveal cross-cutting breakdowns in Alzheimer’s disease
“Single-cell profiling” is helping neuroscientists see how disease affects major brain cell types and identify common, potentially targetable pathways.
MIT engineers grow “perfect” atom-thin materials on industrial silicon wafers
Their technique could allow chip manufacturers to produce next-generation transistors based on materials other than silicon.
Can you trust your quantum simulator?
A new technique helps verify the accuracy of experiments that probe the strange behavior of atomic-scale systems.
Computers that power self-driving cars could be a huge driver of global carbon emissions
Study shows that if autonomous vehicles are widely adopted, hardware efficiency will need to advance rapidly to keep computing-related emissions in check.
Holding information in mind may mean storing it among synapses
Comparing models of working memory with real-world data, MIT researchers find information resides not in persistent neural activity, but in the pattern of its connections.
Study: Extreme heat is changing habits of daily life
New research quantifies how much very hot temperatures restrict outdoor activity in China.
Enzyme “atlas” helps researchers decipher cellular pathways
Biologists have mapped out more than 300 protein kinases and their targets, which they hope could yield new leads for cancer drugs.
Moving water and earth
A new understanding of how particle shape controls grain flow could help engineers manage river restoration and coastal erosion.
Program teaches US Air Force personnel the fundamentals of AI
MIT researchers developed and studied a customized AI training program for users with varied backgrounds, which could be delivered across large organizations.
Research collaboration to examine parent-child learning interactions’ impact on child skill and curiosity
J-PAL North America and the University of Chicago’s Behavioral Insights and Parenting Lab will evaluate two approaches to text-based parental engagement programs that motivate two distinct kinds of learning interactions.
Riddle solved: Why was Roman concrete so durable?
An unexpected ancient manufacturing strategy may hold the key to designing concrete that lasts for millennia.