The way we train AI is fundamentally flawed
The process used to build most of the machine-learning models we use today can’t tell if they will work in the real world or not—and that’s a problem.
by
Will Douglas Heaven archive page
November 18, 2020
“The training process can produce many different models that all pass the test but… these models will differ in small, arbitrary ways… These small, often random, differences… can lead to huge variation in performance in the real world. In other words, the process used to build most machine-learning models today cannot tell which models will work in the real world and which ones won’t.”