Using AI and old reports to understand new medical images
Scientists employ an underused resource — radiology reports that accompany medical images — to improve the interpretive abilities of machine learning algorithms.
How quickly do algorithms improve?
MIT scientists show how fast algorithms are improving across a broad range of examples, demonstrating their critical importance in advancing computing.
SpectrumX collective launches the first NSF Spectrum Innovation Initiative Center
MIT Haystack Observatory will be part of the new radio spectrum management and coordination center.
Research collaboration puts climate-resilient crops in sight
MIT professors Dave Des Marais and Caroline Uhler combine plant biology and machine learning to identify genetic roots of plant responses to environmental stress.
Making self-driving cars safer through keener robot perception
PhD student Heng Yang is developing algorithms to help driverless vehicles quickly and accurately assess their surroundings.
A universal system for decoding any type of data sent across a network
New chip eliminates the need for specific decoding hardware, could boost efficiency of gaming systems, 5G networks, the internet of things, and more.
Q&A: Dina Katabi on a “smart” home with actual intelligence
MIT professor is designing the next generation of smart wireless devices that will sit in the background, gathering and interpreting data, rather than being worn on the body.
Using adversarial attacks to refine molecular energy predictions
MIT researchers find a new way to quantify the uncertainty in molecular energies predicted by neural networks.
Who can bend light for cheaper internet?
ARROW, a reconfigurable fiber optics network developed at MIT, aims to take on the end of Moore’s law.
Last-mile routing research challenge awards $175,000 to three winning teams
Competing research teams trained machine learning models to predict optimal routing based on real field datasets.
System trains drones to fly around obstacles at high speeds
New algorithm could enable fast, nimble drones for time-critical operations such as search and rescue.
Exact symbolic artificial intelligence for faster, better assessment of AI fairness
Probabilistic programming language allows for fast, error-free answers to hard AI problems, including fairness.
Geologists take Earth’s inner temperature using erupted sea glass
The results could help scientists unravel the processes underlying plate tectonics.
Lincoln Laboratory convenes top network scientists for Graph Exploitation Symposium
Researchers share progress applying network science to disinformation tracing, Covid-19 modeling, and machine learning.
Making it through a pandemic senior year with curiosity and community
Miles Johnson ’21, a recent graduate in mathematics and EECS, employed a strong dorm network and personal interests including rock climbing and jazz to complete a rich MIT experience.