“Liquid” machine-learning system adapts to changing conditions
My list

MIT-liquid-networks-01Hero-press

MIT researchers have developed a type of neural network that learns on the job, not just during its training phase. These flexible algorithms, dubbed “liquid” networks, change their underlying equations to continuously adapt to new data inputs. The advance could aid decision making based on data streams that change over time, including those involved in medical diagnosis and autonomous driving.

“This is a way forward for the future of robot control, natural language processing, video processing — any form of time series data processing,” says Ramin Hasani, the study’s lead author. “The potential is really significant.”

+INFO: MIT

Related Posts

Only accessible for registered users
This content is available only for registered users
TO: $$toName$$
SUBJECT: Message from $$fromName$$