Machine Learning

Ten years ago, a title like this would get a ton of clicks. The reason why – to see what type of sorcery this person is referring to. Machines learning? Computers sitting in school, taking classes next to my kid? What the heck is this person talking about?

Today, however, it's highly likely most have heard of machine learning. Yet, there is a lot of "learning" going on in the tech sphere. We also have artificial intelligence (AI) on top of all this. The terms overlap and can be confusing to separate. But we promise they're not. Machine learning is comprised of algorithms. These are responsible for most of the advancements in the larger artificial intelligence sector that you see and read about. These fancy algorithms use statistics to flesh out patterns. Giant data sets are uploaded, which encompass everything from images to numbers and words. The set is then fed into the algorithm, and identified patterns are pushed out.

The first important distinction to make is between AI and machine learning. AI is the broader concept of machines, in general, carrying out tasks, while machine learning is the practical application of the algorithm to come up with the patterns. The first big breakthrough that led to machine learning driving AI forward was in 1959. Arthur Samuel was an academician and researcher and began experimenting with the idea of teaching machines to learn for themselves. The second breakthrough came with the Internet. For the first time ever, we had digital information that was stored and primed to be analyzed.

What followed was then the development of neural networks. These are computer systems that classify information in a remarkably similar way as the human brain. The system can be programmed and taught to recognize images and patterns and then categorize the data accordingly. Machine learning has evolved to now being able to read text, and even determine whether the person writing is angry, congratulatory, or expressing any range of emotion. Machine learning powers systems like Alexa and Siri, as well as Google search engines and even Spotify, YouTube, and Netflix. Data is collected about you in each of these instances, and the machine makes a guess (a highly educated one) as to what you'll like next.

While some have dubbed machine learning and AI as creepy and intrusive, the ease at which we operate online is thanks to these two concepts. Yes, it's true, talking with your partner in the evening about bedspreads and then waking up the next day only to be bombarded with bedspread advertisements can be annoying. But you know what else – it's also what you were planning on purchasing, so now there's some items to ponder.

Machine learning isn't going anywhere. It's akin to discovering gold. Once those shiny minerals were uncovered and subsequently coveted, people didn't weigh the option of not continuing to mine for gold. It was too valuable, and the demand is still through the roof. Machine learning is the same thing. The amount of good it brings outweighs the negatives. There are naysayers, but even they secretly understand the value and take advantage of like the rest of us.