MIT’s new ‘liquid’ neural network learns on the job — so robots can adapt to changing conditions

Jan29
by on January 29, 2021 at 4:35 pm
Posted In: Insider


MIT researchers have invented an adaptive “liquid” neural network that could improve decision-making in self-driving cars and medical diagnosis. The algorithm adjusts to changes experienced by real-world systems by changing their underlying equations as they receive new data. “This is a way forward for the future of robot control, natural language processing, video processing — any form of time series data processing,” said Ramin Hasani, the study paper’s lead author. “The potential is really significant.” [Read: How this company leveraged AI to become the Netflix of Finland] Hasani said the system is inspired by a tiny worm — the C. elegans: It…

This story continues at The Next Web

└ Tags: , , , , , , , ,
 Comment 

Scientists develop AI-powered ‘puff technology’ to help smokers quit

Jan28
by on January 28, 2021 at 5:16 pm
Posted In: Insider


Scientists are developing an AI device that helps people quit smoking by predicting when they’ll crave a nicotine hit. The “Level” system is an onboard computer for e-cigarettes that precisely monitors vaping behavior. It’s designed to help users control their cravings or reduce the amount of nicotine they’re inhaling. Level will incorporate insights from an observational study launched today by the University of East Anglia. The research team has developed a small monitoring device that screws onto tank-based e-cigs. It then measures the number and duration of a user’s vapes, the time between each drag, and the power used per puff. This data will create a personalized…

This story continues at The Next Web

└ Tags: , ,
 Comment