Why Enter the Machine Learning Industry?

 

We'll look at why you're interested in machine learning in this post.

We'll look at some questions that can help you get to the bottom of why you're interested in the field.

We will conclude with a map depicting the four main "whys" so that you can determine where you fit and what resources to target.

Question Your Why

What piques your interest in machine learning? Have you given this question much thought?

Knowing you're why is beneficial because you can use it as a filter to select projects and tasks that you enjoy working on. If you can't think of a clear why that can be useful as well because it can motivate you to try a variety of things to figure out what you like or want to do.

You may be interested in machine learning for a variety of reasons. Perhaps you're reacting to articles in the media and on the news about big data and data science. You may have caught a glimpse of machine learning in a tool or from a friend and thought it was cool. There could be many reasons for this, but learning machine learning is difficult. To have the confidence and perseverance to get through the difficult and frustrating parts of studying, you must have a strong why to fall back on.

I'm going to pose some rhetorical questions, and I want you to consider (even write down) your responses and see which one resonates the most with you. Keep an open mind; one question is not better than another.

What do you want machine learning to do for you?

Fix a Problem

Do you believe machine learning can help you solve a problem?

Perhaps it is an ongoing business issue or a workplace issue. Perhaps there is an opportunity in the market. Nonetheless, you consider machine learning to be a tool for you to learn and apply to a problem.

In this case, you might be interested in learning about tools that provide quick implementations of algorithms. You'll also be interested in innovative applications of these tools, such as case studies on problems similar to the one you want to solve.

Technical Excellence

Is learning machine learning a sign of success?

Maybe machine learning is a popular technical field, and learning new and difficult technologies and tools gives you a lot of pride. Perhaps you see machine learning as your next big challenge and opportunity for growth, as well as an opportunity to demonstrate your ability to learn and master technical materials.

If this describes you, you might be interested in algorithm books, which allow you to quickly learn a method and how to apply it without having to delve into the most recent research. You will also most likely be interested in taking courses, competing, and developing your own algorithms.

What do you hope to achieve with machine learning?

Extend the Field

Do you already have some experience with machine learning and want to broaden your horizons?

Perhaps you've dabbled in machine learning and read a book or completed a course. You've discovered a question or a method that you just can't put down, and you want to not only go deep on it, but you also want to push the boundaries of what that method can and has been shown to be capable of.

If this sounds familiar, you might be interested in in-depth subject matter such as research papers and monographs. You might also be interested in hearing expert opinions on the subject and where the frontiers are.

Do what was previously thought to be impossible.

You have some machine learning experience and domain expertise, and you want to do things in your domain that would be impossible without machine learning. These are not necessarily problems like those mentioned above in the "Solved Problem" section, but rather the extension of a domain using machine learning's experience and capabilities.

You'll be interested in methodologies ranging from data mining to pattern recognition. You'll also be interested in case studies of machine learning methods' discoveries and extensions in similar domains.

Machine Learning Map

This is an oversimplification of the field, but we can categorize our motivation to learn machine learning based on the type of work we want to do. We can divide the work we want to do into two categories: problem-solving in machine learning and problem-solving in another domain. Tasks can be divided into two categories: practitioner tasks and researcher tasks.

I attempted to summarise this information in a table, which is shown below.

The table is divided into two domains: machine learning and the other domain (such as analytical chemistry, petroleum mining or transport analysis.). The table is divided into two columns based on role: practitioner and researcher. Each box indicates the type of task for that domain-role intersection, which is either to solve a problem or to broaden the field. And each cell in the table lists the different types of resources that a person interested in that task might find useful.

Each cell can be thought of as a reason why you want to learn more about machine learning, and the list of resources can assist you in that endeavor.

This is only one way to cut the pie, but I've been thinking about it for a few weeks. I worked hard on the groupings and would love to hear what you think of them; please leave a comment. I'd love to get some experts to start poking holes in it so we can see the model's strengths and limitations (all models are wrong to some extent).

Please leave a comment and tell me where you are why fits in and with what you identify.