SCIENCE SPIN-OFFS

[dockzef]

High Dimensional Human Guided Machine Learning

Eric Holloway, Robert Marks II

(Submitted on 4 Sep 2016)

Have you ever looked at a machine learning classification model and thought, I could have made that? Well, that is what we test in this project, comparing XGBoost trained on human engineered features to training directly on data. The human engineered features do not outperform XGBoost trained di- rectly on the data, but they are comparable. This project con- tributes a novel method for utilizing human created classifi- cation models on high dimensional datasets.

http://arxiv.org/abs/1609.00904

David I have been looking for some time for this type of proof of the AI getting even with Human intelligence. My prediction from Ray Kurtz. info was that this would have happened in 2015 from back in 1999. Either way it does not pay to make predictions so it represents the on ramp to the superhighway of information overload with the machines guiding us to where we need to be to learn at an ever increasing rate... Of course until we figure it all out! LOL
.