Here in Harrod’s Creek, Kentucky there is not too much chatter about machine learning. It is hunting season. Time to get out the Barrett Automatic Rifle and go hunting for varmints. Sundown yesterday when calm returned to the hollow, I read “Machine-Learning Maestro Michael Jordan on the Delusions of Big Data and Other Huge Engineering Efforts.”
My thought after reading the IEEE article was that I was really tired of the artificial intelligence yap yap. Now a whiz at UCal Berkeley is pointing out that some of the methods are a “cartoon.” The Dr. Michael Jordan says:
I think data analysis can deliver inferences at certain levels of quality. But we have to be clear about what levels of quality. We have to have error bars around all our predictions. That is something that’s missing in much of the current machine learning literature.,,if people use data and inferences they can make with the data without any concern about error bars, about heterogeneity, about noisy data, about the sampling pattern, about all the kinds of things that you have to be serious about if you’re an engineer and a statistician—then you will make lots of predictions, and there’s a good chance that you will occasionally solve some real interesting problems. But you will occasionally have some disastrously bad decisions. And you won’t know the difference a priori. You will just produce these outputs and hope for the best. And so that’s where we are currently.
In short, marketing hyperbole takes precedence over the plodding realities of the steps required of a person aspiring to a PhD in statistics is supposed to follow.
With regard to the applications that deliver predictive outputs, Dr. Jordan says:
But unless you’re actually doing the full-scale engineering statistical analysis to provide some error bars and quantify the errors, it’s gambling. It’s better than just gambling without data. That’s pure roulette. This is kind of partial roulette.
I strongly recommend you read the interview. I would not involve a search or content processing marketer in the exercise, however.
Stephen E Arnold, October 24, 2014