Credit: Data Science Central
Leveraging the abbreviation “vs” in and of itself begins to stir the insides of the ever-faithful, “until death do we part” neural network enthusiasts because, lets face it, they are riding a wave that is driving the stock prices in both the private commercial and public commercials sectors. Most of the applications talked about today which leverage the all-to-mysterious but oh-so-exciting “AI” or “Artificial Intelligence” are implementing supervised learning approaches to solve their problems. We should be careful, however, not to overfit these tools to problems that ought be solved otherwise. Neural nets have experience a hype-curve before in which they overpromised, underdelivered, and almost died. So, let’s not ruin a good thing before it gets REALLY GOOD.
I believe we need a slight paradigm shift in the way we see these technologies as being at odds with one another, AND we need an education on the differences in function and use-cases best suited for one, the other, or a combination of the two (yes, this is entirely possible). Over the past 5 years, after hundreds of conversations with executives and engineers in dozens of household named companies, I have come to realize that most people, although talking extensively on the topic and spending thousands…..millions on some of these solutions, don’t actually understand them well at all.
This is my first blog post, and I don’t have any followers or listeners at this point, so to test how this all works…and if this is even worth my continued effort, I am going to leave this here and see if I get any views or responses.