Can machine-learning models overcome biased datasets?
A model’s ability to generalize is influenced by both the diversity of the data and the way the model is trained, researchers report.
A model’s ability to generalize is influenced by both the diversity of the data and the way the model is trained, researchers report.
The planet’s night side likely hosts iron clouds, titanium rain, and winds that dwarf Earth’s jetstream.
Historian Tanalís Padilla’s new book about activist rural schools in Mexico highlights long-running tensions in the nation’s politics.
Engineers build a lower-energy chip that can prevent hackers from extracting hidden information from a smart device.
The advance may enable real-time imaging devices that are smaller, cheaper, and more robust than other systems.
A new study sheds light on how a protein pumps toxic molecules out of bacterial cells.
Lincoln Laboratory leads a large-scale measurement campaign in New York City to improve air dispersion models and emergency protocols.
Different types of these branch-like projections process incoming information in different ways before sending it to the body of the neuron.
MIT researchers find activating a specific acetylcholine receptor in the brain reduces cocaine use in rodents.
A new analysis offers guidance on the size of nanoparticles that could be most effective at stopping internal bleeding.
A new deep-learning algorithm trained to optimize doses of propofol to maintain unconsciousness during general anesthesia could augment patient monitoring.
Researchers with the KATRIN experiment determine that neutrinos are lighter than 0.8 eV/c2.
Passive solar evaporation system could be used to clean wastewater, provide potable water, or sterilize medical tools in off-grid areas.
Single-cell gene expression analyses of human cerebrovascular cells can help reveal new drug targets for Huntington’s disease.
A subset of the finalists will be announced as multiyear flagship projects this spring.