Google is a step further towards developing an advanced AI architecture called Pathways - a single model that can learn to do multiple tasks that are currently accomplished by employing multiple algorithms.
This new technology is called LIMoEa - an acronym that stands for Learning Multiple Modalities with One Sparse Mixture-of-Experts Model. It’s a model that processes vision and text together.
There are currently other architectures that do similar things, but Google's breakthrough is in the way their new model accomplishes these tasks, using a neural network technique called a Sparse Model.
LIMoE works on processing text and images simultaneously - lowering the computational cost and therefore making the model more efficient and as such, we could be on the verge of something quite exciting.
Let's see!
Google announced a new technology called LIMoE that it says represents a step toward reaching Google’s goal of an AI architecture called Pathways. Pathways is an AI architecture that is a single model that can learn to do multiple tasks that are currently accomplished by employing multiple algorithms. LIMoE is an acronym that stands for Learning Multiple Modalities with One Sparse Mixture-of-Experts Model. It’s a model that processes vision and text together.
https://www.searchenginejournal.com/google-limoe/454673/#close
unknownx500
