The Thing Holding AI Back Is the Thing It Needs Most, Data
November 30, 2017
Here’s an interesting problem: for artificial intelligence and machine learning to thrive, it needs a massive amount of information. However, they need so much data that it causes hiccups in the system. Google has a really interesting solution to this problem, as we learned in the Reuter’s article, “Google’s Hinton Outlines New AI Advance That Requires Less Data.”
The bundling of neurons working together to determine both whether a feature is present and its characteristics also means the system should require less data to make its predictions.
The leader of Google Brain said, “The hope is that maybe we might require less data to learn good classifiers of objects, because they have this ability of generalizing to unseen perspectives or configurations of images.
Less data for big data? It’s just crazy enough to work. In fact, some of the brightest minds in the business are trying to, as ComputerWorld said, “do less with more.” The piece focuses on Fuzzy LogiX and their attempts to do exactly what Google is hypothetically saying. It will be interesting to see what happens, but we are betting on technology cracking this nut.
Patrick Roland, November 30, 2017