Teaching language models grammar really does make them smarter
Researchers submit deep learning models to a set of psychology tests to see which ones grasp key linguistic rules.
Researchers submit deep learning models to a set of psychology tests to see which ones grasp key linguistic rules.
Signals help neural network identify objects by touch; system could aid robotics and prosthetics design.
MIT Quest for Intelligence-sponsored undergraduate research projects demystify AI.
Autonomous control system “learns” to use simple maps and image data to navigate new, complex routes.
CSAIL system can mirror a user's motions and follow nonverbal commands by monitoring arm muscles.
“Metasurfaces” that manipulate light at tiny scales could find uses in cellphone lenses, smart-car sensors, and optical fibers.
New program will focus on rapid deployment of artificial intelligence innovations in operations, disaster response, and medical readiness.
New tools can find patterns in vast online data to track and identify users on illicit forums.
Third annual MIT Statistics and Data Science conference gathers a growing interdisciplinary community of researchers and practitioners.
New method quickly detects instances when neural networks make mistakes they shouldn’t.
In some cases, radio frequency signals may be more useful for caregivers than cameras or other data-collection methods.
MIT CSAIL project shows the neural nets we typically train contain smaller “subnetworks” that can learn just as well, and often faster.
Algorithm stitches multiple datasets into a single “panorama,” which could provide new insights for medical and biological studies.
Data-sampling method makes “sketches” of unwieldy biological datasets while still capturing the full diversity of cell types.
Researchers unveil a tool for making compressed deep learning models less vulnerable to attack.