How to be successful with self learning?

Leonardo did not invent the scientific method, nor did Aristotle or Alhazen or Galileo or any Bacon. But his uncanny abilities to engage in the dialogue between experience and theory made him a prime example of how acute observations, fanatic curiosity, experimental testing, a willingness to question dogma, and the ability to discern patterns across disciplines can lead to great leaps in human understanding.
Walter Isaacson, Leonardo Da Vinci

If you are confused, frustrated, alone and in a tremendous self-pressure, you are on the right track for self-learning. You don’t even need to read this blog (at all). Just continue what you are doing. But if you are at the beginning of a chaos, and have no idea where to start, or how to learn; I promise you that this will help you. A little bit of chaos incoming.

To give you a context, this blog is for those who are learning something new in their college years, or later in their career. When the formal education system has failed them or when they are no longer able (/want) to go into that system. This blog is for those who want to learn but don’t have a guide or direction, that’s it!

Without further ado, here are 5 points, short and simple.

Choose the Simplest-Smallest-Achievable Goal

At the beginning of any journey, many struggle to get started! Most of the time, when I talk to people, they ask me, Where to start with ML?. Then I suggest to them some courses online. After a few months, they will ask me the same question!

It doesn’t matter if you are a total beginner or in the middle of your journey. These questions will always arrive. What to do next? Where to start? And the simplest, but harder-to-implement answer is to find the most basic, most mundane-looking source you can find. If you are learning ML, find the most basic or most well-known course. Choose the simplest-smallest goal that you can think of.

To choose that goal however, you would have to follow your intuition. In the beginning, most of us get overwhelmed by the amount of knowledge there to learn. Learning how to divide the goal into simpler, smaller, achievable goals is the most important skill you would develop in this journey.

Not to forget, your goal could be wrong (which is more often than you would like). But this is just a part of the journey. Once you know what’s wrong, you would learn to set up correct goals too!

Avoid Rabbit Hole

There’s tons and tons of knowledge on every subject. Machine learning is based on statistics, probability theory, linear algebra, and graph theory… the list is pretty long. Each of these subjects have their entire department in the universities for the last hundreds of years. So you can’t learn any subject or even a concept ‘in depth’.

It is an ocean without a base. But our curiosity points us in that direction, making us dive more and more. I have wasted my weeks diverting from the main goal and digging into something completely random.

The solution for that is to create a simple chart. First, you would ask yourself what I want to learn. Then when you have read the concept/ consumed the resource. Just ask yourself that same question, was that enough to satisfy my answer? If yes, just move on. Even if you are unsatisfied with the depth, you would still have the width that you wanted in the first place.

Learn to ask the right questions

When I am getting some recognition as a professional (lol), sometimes people ask me to help them with their code or the concept they are struggling with. I am not saying I am the expert, but I can see that they lack something which I used to lack too. They don’t know how to ask good questions!

This is a common thing among the Indians I would say. Our education system has been a spoon-feeding system. When we have certain doubt, just ask the teacher and they will answer it for you. But when we get to that point where we don’t have anyone to teach us, this mindset completely fails.

If you ask, what I call, the primary doubt to someone they will most likely (if good-hearted) will resolve your doubt, but then you will lose a treasure they have inside them, forever. Let me elaborate.

If you have a certain doubt or a certain query, instead of jumping into someone’s chat box, just acquire as much knowledge as you can around that doubt. Acquire the entire area surrounding that little hill. You know you can’t go to the top of that hill yet, but you would still know the surrounding area at least. I call this secondary (or above) doubt. When you have this knowledge, feel free to ask. Most of the time, the person will be more comfortable talking to you as you have (at least) keywords/ jargon to talk about the concept.

Secondary doubts make that student-guide bond stronger too. This will help you to build long-term relationships.

In the world of ChatGPT, secondary thoughts will make you better at prompt engineering too.

Learn to develop a thick skin

This sounds a little cheesy but it is true. In the self-learning journey, most of the time you are going to wander in the wrong direction or very little progress in weeks or months! This is just normal.

Remember one thing, Don’t Give Up!

Even if you fail there’s nobody to judge you or nobody to know. Often, the self-learning journey is lonely. So people won’t even notice you if you fail. Don’t take yourself too seriously. If you fail, just try the other way. If you are stuck, ask for help. Self-learning is not about how much you can do, is about how far you can go. Okay, that was way too cheesy a paragraph. Let’s move on.

Paradox of feedback

In any formal education system, you have end-year/semester exams to give you feedback on what you have done. If you score well, it will satisfy you to do a party with your friends (it doesn’t matter even if you have not understood anything). But in a self-learning journey, there’s no such exam, no such goal, no medal to achieve, no exam to crack! Nothing, at max, you can get a new flashy job, but that too won’t satisfy you (I bet).

On the other hand, many things can give you negative feedback. Like Einstein famously said, “You can’t judge a fish by its ability to climb a tree”. The entire world would judge you based on your ability to climb the tree, that’s just the how world works, especially if you are finding a new job.

So what to do? How to keep that motivation alive when there’s nothing that says You did good. The only answer I can think of is to journal your progress. Each week, each day or if possible each hour. Just put down what was your goal for that week, what were your doubts, how you solved it and what was the answer. When you will look at this after a few months, you will understand your progress better.

Do it like nobody’s watching, because nobody’s watching. 😉

Conclusion

I am a big believer in self-learning. I have struggled with formal education all my life. That’s because formal education, although somewhat effective, is designed to give us answers to our questions. It never teaches us to be more curious or to be more vibrant.

I believe the self-learning journey is a wholesome package to live a better life. You will be forced to follow intuition, to believe in your mind and curiosity. It will teach you how to not give a fuck about others, how to create your own happiness, your own treats. You would have to put in a tremendous amount of time, a ton more than just 10,000 hrs to learn and master things. So you would have to be consistent and disciplined. All of these qualities, which we are listing down are so-called success mantras for life.

But self-learning teaches you to believe in the journey not to focus on just the goal. Life is a journey (not even a marathon).

I am referring You, to the reader of this sentence.

I don’t want to oversell self-learning. Thanks for reading.