Is Machine Learning for Me? 10 Self-Examination Questions

We are talking 97 million jobs.

That’s how high the demand for specialists in machine learning and artificial intelligence-related industries is predicted to expand by 2025, according to a report by the World Economic Forum.

Obviously, the market can spare one of these job places for you, if only you’re 100% sure that machine learning fits like a glove, especially if you want to “wear” it as a long and happy career.

Before asking some self-checking questions and giving answers to yourself to understand that machine learning is clearly for you, start with the zero-question and see how well you can decipher between three buzzwords.

Question #0. What’s Machine Learning vs Artificial Intelligence vs Data Science?

What’s Machine Learning (ML)?

Machine learning is a subset of artificial intelligence. It presupposes that an ML engineer feeds machines with data and “teaches” them to understand it and make decisions based on some patterns without the slightest human intervention.

What’s Artificial Intelligence (AI)?

Artificial intelligence implies machine-guided knowledge adjustments based on inputs and the very decision-making process.

What’s Data Science (DS)?

As a matter of fact, everything starts with DS and it incorporates both ML & AI.

Data science is an interdisciplinary field that deals with various methods, tools and techniques to work with volumes of data and extract meaningful insights from it.

Data, statistical concepts and calculations are fundamentals of ML. They are essential pillars you cannot simply bypass or omit, with so many still-to-uncover secrets in the underwater depths of artificial intelligence.

Image Source: Data Science Vs Machine Learning (author: Hema Sri Kovela)

So, figuratively, jumping into machine learning, you jump into the data ocean with algorithms and patterns.

Could you swim in it well?

Let’s see answering ten self-examination questions below together to make sure machine learning is undeniably for you.

Question #1. Am I ready to dive into data, data, and more data + statistical concepts?

If such phrases as “linear regression” and “probability density function” don’t scare you away, it’s ok then. We may continue.

Question #2. What’s the average salary of a machine learning specialist?

Let’s talk money.

Note: salaries in machine learning differ dramatically from country to country and city to city. You can also check the highest and lowest paying cities and companies too. Don’t forget about other factors that will influence your pay: years of experience, education, professional associations & certifications, company’s status, etc.

(Pst, the gender pay gap is, unfortunately, still alive.)

The average base salary of an ML engineer in the United States, for example, is ~ $100,000–110,000 per year, with the lowest number – $71,000 and the highest one – $300,000.

Do these numbers satisfy your objectives? Move on to find out what means to resort to, if you’d like to get a high-paid job in ML and how much time to spend on preparation.

Question #3. What competencies and skills do I need to become a data scientist in ML?

Here’s what knowledge-set is normally expected from machine learning engineers:

• Computer science and programming
• Probability and statistics
• Software engineering
• Applied mathematics
• Machine learning algorithms (APIs, packages and libraries like TensorFlow, Sci-Kit learn, Spark MLib, Theano, Keras, etc.)
• Dataset structures, modeling and model evaluation
• Practical and hands-on experience in statistics, coding, data analytics and the like

Have a closer look at the traditional ML syllabus for the bottom-up approach.

Now to the top soft skills required for an ML specialist position:

• Persistence
• Communication and presentation skills
• Team-player’s skills
• Thirst for never-ending improvement and learning
• Problem-solving skills
• Time management

Now answer some extra questions: do the above-mentioned points comply with your expertise and skill set and are you willing to master/develop those? Like, for instance, consider putting a programming language or two under your belt.

Question #4. What are the basic programming languages a machine learning expert should know?

Mastering Python and Pandas Python, in particular, is a must-do for the background machine learning skills.

Depending on your needs, however, you might want to select from the best programming languages for DS (besides Python) such as R, Java, SQL or Julia. Alternatively, those who’re way ahead of the beginner’s level and are aiming at complex high-level algorithms may go with Scala.

Question #5. How can I get prepared for a career in ML?

What can you do to “groom” yourself properly for landing one of the dream jobs in machine learning?

Delve into these ways:

1. Peer into the forecasts in this niche

What if AI robots replace us, steal our jobs and make us slaves? Just kidding.

Don’t worry, your job in data science is safe and, more importantly, you can be the one to teach them how to live peacefully with people.

So, quantum supremacy has been achieved and it can define the future of data science and AI. Contemplate on other artificial intelligence & ML predictions to know what you’re stepping into.

2. Research the ML developer job market

By now, you already know how much you can earn as a machine learning engineer (we’ve covered it by answering Question #2).

Now to the job trends, openings, top tech companies-employers, career prospects, etc.

The AI job market is on fire. That’s true. You should know some peculiarities, however. E.g., jobs in AI cybersecurity are fewer and harder to find. Besides, a decent computer science background is practically always preferred to land such a career.

3. Outline your desired job role and responsibilities

How do you imagine your working day? (Throwing a ping-pong ball at the wall doesn’t count.)

Determining your job duties, tasks and requirements is the first step to take while preparing for a data science interview.

4. Join dedicated communities for DS practitioners and professionals

Settle down on Kaggle and Github. Both can help you get an entry level data science job and demonstrate your experience clearly. Start building a portfolio with the simplest test assignments and back up your resume projects. E.g., on Kaggle, it’s possible to explore and build models, work with other ML developers or even enter competitions to solve specific challenges.

5. Boost your data science skills continuously

(Your ping-pong ball throwing skills are great, I’m aware of that.)

The toughest part of developing a new competence and skill is knowing how exactly and where to get started. Read on to receive an answer.

Question #6. What should I start with to try myself in ML?

Did you know that 59% of employed data scientists learned skills via massive open online courses (MOOCs) or on their own?

In fact, you may get a world-class machine learning education without paying a dime. 

Start either with short steps and cover each topic related to ML one at a time or immerse into all-inclusive programs.

One-by-one approach

Courses:

Books (from those for a complete noob to some “heavier” ones):

• Machine Learning For Absolute Beginners by Oliver Theobald
• An Introduction to Statistical Learning by Gareth James and others
• Python Machine Learning by Sebastien Raschka
• Hands-On Machine Learning with Scikit-Learn, Keras, and TensorFlow by Geron Aurelien
• Machine Learning for Hackers by Drew Conway & John Myles White
• Deep Learning (Adaptive Computation and Machine Learning series) by Ian Goodfellow, Aaron Courville & Yoshua Bengio

Top 3 resources:

Grab some more resources for learning AI, ML and data science.

All-in-one approach

You can swallow book by book, try website by website and dedicate your time to all-encompassing machine learning courses & certifications simultaneously to speed up the process.

Question #7. How much time does it take to learn machine learning?

Scroll up a bit, return to the last step in Question #5 and look at the last word.

If you haven’t done that, here it is: continuously.

That’s how you should approach the process of learning. Because how can you possibly grasp data science and machine learning from A to Z literally? When AI advancements accelerate with an enormous speed and every second there appears something new in this continuum.

Well, what else could you expect from a quantum processor that performs a complicated task in 200 seconds in comparison to 10,000 years a world’s supercomputer would need to do that?

Still, even though it depends individually, let’s estimate an approximate duration to bite and then gnaw the granite of this science.

10 days may be enough, as a rule, to just get acquainted with the nitty-gritty theory and practice.

Have you already said “hi” to ML and data science and “tasted” some statistics earlier?

Great. You’ll spend comparatively less time on digging deeper into this realm, as you’ve got some firm footing by this time.

Certificate programs and bootcamps in AI & ML typically take 4–12 months. By and large, 6 months is an average timespan for studying machine learning.

In reality, if you decide to go with an approach Madhukar Jha suggests based on his experience, for example, you might need some more time (10 months, to be exact).

Image Source: Step-by-Step Process of How I Became a Machine Learning Expert in 10 Months (author: Madhukar Jha)

Picking web-based courses, you should pay attention to their duration as well, as it’s a crucial factor, because longevity matters for a successful completion. MOOCs face an astronomical dropout rate of 96%.

Bet you’re a toughie and can survive any course till the end.

Question #8. How easily do I get discouraged and leave everything, if my first idea doesn’t work?

Can you fly off the handle and throw your mouse or keyboard out of the window? 

Is it a normal thing for you to do, when something goes wrong? And then you may say something like “Oh, snap” (the mildest version), sack this and do a runner, right?

Nuh-uh. Not with machine learning tasks.

“Success takes longer than you think” – Scott H. Young

You can’t expect to translate your idea into a list of operations straightforwardly. 

Most frequently, things don’t work from the get-go and may break in the second part. Persistence and patience should be the two pals to accompany you during each task in ML.

It’s the winning combination to feed your motivation too. So, here comes the one but last question.

Question #9. What’s my motivation to do machine learning?

Well, my friend, I can’t help you much with this one, because only you can feel your inner state and desires.

As sexy and prestigious as it may sound, you shouldn’t get deluded by a mere phrase “machine learning expert”. Instead, be honest with yourself and determine what drives you to pick this career.

If you’re motivated enough, the last question (and answer) won’t shatter your inspiration and goals.

Question #10. What are the most widespread concerns & ethical bias in AI and ML?

Remember those robots (from Question #5) that may ignite an AI invasion, steal the jobs and blah-blah-blah?

Some people do consider engineers who teach machines to be the drivers of the apocalypse. No wonder such concerns mount, as AI takes a bigger decision-making role in more and more industries.

You should also comprehend the basics of machine morality and ethics in AI.

Are you fully aware of the ethical implications of implementing ML-driven solutions? 

These are mostly related to gender-biased algorithms, potential racial bias, risk assessment pitfalls in the criminal justice system, etc.

Btw, it’s one of the questions machine learning engineers can expect in a job interview.

Everyone is into Machine Learning. Can It Be Your Choice Too?

Let’s face it – nothing is specifically designed for you. Even the gloves (unless you ordered a unique design for those). 

However, when it comes to your career path, possibly in machine learning, as the topic goes, you should reflect on your wants, needs, abilities, etc., do some self-analysis and identify how well you can harmonize with that environment.

And here you go – you’ve already answered ten questions to test and examine yourself and weigh all the variants for a career in data science and ML.

You’re also aware now that buzz and hype around machine learning isn’t without a reason.

Because it’s such a delicious data-full dessert! 

Particularly for those who never mind getting lost in data labyrinths and coding or those innovation evangelists who want to be world-reshapers. Or both.

So grab your dessert spoon, if you’re ready, and on you go.