One of the core principles of adult learning theory is that adult learners bring with them a wealth of experiences that can be accessed and used when learning new concepts and techniques. In traditional adult learning settings, this prior knowledge can often be a great support for the construct of learning, and can even enhance and help guide examples and analogies that might be used by them.
However, prior learning can also be a burden, a wall if you will, when it is based on a litany of business articles and books pontificating on the rise of the machine. While subject-matter expert opinion can often provide guidance for how a new technology can be used to build value for an organization, the differing perspectives on new technologies, particularly something as complex as Artificial Intelligence and machine learning, can sometimes lead those new to the subjects down “rabbit holes of doom.”
As such, it’s important for learners to keep in mind that some connections to prior learning aren’t necessarily accurate ones just because someone wrote an article about it. Take the idea of machine learning algorithms as used in natural language processing versus computer vision. Both processes take similar approaches to classifying an object based on its features, whether those be words or pixels, respectively. However, the implementation of these seemingly similar approaches can be vastly different, often combining or stacking machine learning techniques to achieve very different outcomes.
Someone new to learning about these concepts may try to take a concept they learned in a data analytics course, for example, and apply it directly to the idea of natural language processing. After all, both are using machine learning to create some understanding. They may even be using the same generally-named algorithm to develop insight, so this knowledge must be transferrable, right?
The problem is that there are likely vast differences in how they are implemented given a variety of factors. One may use unstructured data to uncover hidden connections while the other may focus on structured data, labelled in such a way that the AI discovers connections amongst features and then categorizes that learning into labelled buckets.
Computers are still essentially 0s and 1s. Yes, some can do amazing things like learn a language and understand humans, compose music, or even write their own code, but deep down it’s still just 0s and 1s.
In other words, knowledge transfer within AI and machine learning is not as simplistic as learning a skill like reading and then just applying it to the act of reading books on different subject matter. If anything, it’s more akin to learning the act of reading in English and then trying to use that knowledge to reading hieroglyphics. In this regard, reading isn’t quite reading after all, and the knowledge gained from reading AI and machine learning articles or books, while useful in the holistic sense of learning, does not necessarily provide the requisite knowledge for truly understanding how these different techniques are implemented within any one category of AI.
So what is an adult learner to do?
Step One: Keep an open mind.
Don’t rush to categorize how an AI works given prior reading, learning, or experience. Observe what the AI does and does not do, even if it is not doing exactly what you anticipated it would do.
Step Two: Remember it’s a machine.
Computers are still essentially 0s and 1s. Yes, some can do amazing things like learn a language and understand humans, compose music, or even write their own code, but deep down it’s still just 0s and 1s. As long as there is that restriction on true understanding, there will likely be limitations on what that AI can do.
Step Three: Recognize the journey.
While the idea of machine learning has been around for decades, recent advances in processing speed, memory capacity, and data availability have brought us to new frontiers of possibility.
The opportunity to seize the value of integrating AI into our businesses has never been more conceivable. But the understanding of the technology itself is only now achievable as long as we patiently and thoughtfully explore what it can do today and where it is likely to go tomorrow.
We are all at the brink of something wonderful and powerful, but it must be remembered that this is just the beginning of the learning journey.