Groups keyboard shortcuts have been updated
Dismiss
See shortcuts
Liquid Neural Networks, A New Idea That Allows AI To Learn Even After Training
16 views
Skip to first unread message
Alan Timm
Apr 20, 2024, 12:53:32 PM4/20/24
to RSSC-List
A fascinating glimpse into what may lie next with neural networks. Daniela Rus and her team is adapting the lessons learned from c-elegans to implement a new generation of neural networks that can continue to learn after training.
Everything that's happening today is based on a highly simplified artificial neuron model from the 1950s. Her team proposes using a more advanced neuron model based on studies of c-elegans, a microscopic nematode with only 302 neurons. The results are really promising. This is a really engaging 12 minute talk.
Alan Timm
Oct 2, 2024, 11:42:48 AM10/2/24
to RSSC-List
A quick update, they've spun this off into a company, https://www.liquid.ai
Today, we introduce the first generation of Liquid Foundation Models (LFMs). LFMs are large neural networks built with computational units deeply rooted in the theory of dynamical systems, signal processing, and numerical linear algebra. This unique blend allows us to leverage decades of theoretical advances in these fields in our quest to enable intelligence at every scale. LFMs are general-purpose AI models that can be used to model any kind of sequential data, including video, audio, text, time series, and signals.
Our name “Liquid” pays homage to our roots in dynamic and adaptive learning systems."
And they've trained a few models using this new technique.
"At Liquid AI, we build new methods for designing powerful AI systems over which we have significant control. We design them the same way engineers built engines, cars, and airplanes: from first principles. Our mission is to create best-in-class, intelligent, and efficient systems at every scale – systems designed to process large amounts of sequential multimodal data, to enable advanced reasoning, and to achieve reliable decision-making.
Today, we introduce the first generation of Liquid Foundation Models (LFMs). LFMs are large neural networks built with computational units deeply rooted in the theory of dynamical systems, signal processing, and numerical linear algebra. This unique blend allows us to leverage decades of theoretical advances in these fields in our quest to enable intelligence at every scale. LFMs are general-purpose AI models that can be used to model any kind of sequential data, including video, audio, text, time series, and signals.
Our name “Liquid” pays homage to our roots in dynamic and adaptive learning systems."
Reply all
Reply to author
Forward
0 new messages