I’m going to give you five steps to study machine learning as effectively as possible in this Article.
1. Live a Healthy lifestyle
The first step to studying machine learning effectively, is to live a healthy lifestyle,That means taking care of yourself as a biological learning machine, consisting of both mind and body. It’s important to eliminate what Dr. Daniel Amen the author of use your brain to change your age called amts, or automatic negative thoughts.
These are thoughts that everybody has had at some point, like, I’m not smart enough or I’m too old or my GPU is AMD quality.
Your brain is a supercomputer and your self talk is the program it will run, So program your computer well. In this information age, you have access to every machine learning resource you need, including data, algorithms, computing power and education.
There’s no excuse to say I can’t learn machine learning because of x. Having a positive peer group can help with this.
Find people that encourage and support your learning journey, you are the sum of the five people you spend the most time with.
And if you can’t find a supportive group locally, find your nearest School of AI chapter, It can be just as fulfilling having social interactions online as it is offline.
Also, exercise is super important to learning, Research shows that anything that’s good for your heart is also good for your brain.
People who exercise regularly have better blood flow and oxygen in their brains, And as an inevitable result. perform better on mental acuity tests.
Find a physical activity you actually enjoy like swimming or biking, And that way you’ll stick to it despite the pain and the phrase you are what you eat is very much true.
Try to avoid too much processed food and sugar and drink lots of water, Make this a part of your lifestyle.
Also, getting a good night’s rest is essential because it’s when your brain consolidates short to long term memory, Lastly, information overload is a real thing in today’s notification driven society and can lead to burning out, make sure to take time to disconnect.
Give yourself a break. It’s okay to isolate yourself from the internet to rejuvenate a bit I put my phone on airplane mode at least two hours before I sleep every night and try to have at least 124 hour period a week where I’m completely disconnected in nature.
2. Optimize your learning environment
What does your study space look like? I find I can study more efficiently when I do so in a designated study space. I’m not as productive if I’m learning in the same room where I sleep.
So I go to an office. it’s my designated learning only zone and it’s easier to focus when you’re in a clean environment. So don’t have a cluttered desk, Try to minimize distractions like Black Mirror as well.
Everybody has different preferences on what is considered a distraction. I try and eliminate all sound using noise cancelling headphones, Libraries are always a great place to study since they are quiet, and coffee shops work as well.
3. Personalized learning path for yourself
I’ve now made several learning paths that hundreds of thousands of wizards have participated in whether they’re related to deep learning or machine learning, or general computer science.
The way I designed these curriculums was by first thinking of the topics in AI that interested me the most, then finding relevant resources using Google helped me learn about those topics in order of complexity, I follow my curiosity.
That’s the best way to stick to any learning path as long as it’s curiosity driven. It’s something that you’re actually interested in, you’ll have no problem putting in the time to learn it.
Think of it like a meta curriculum, your curriculum can can consist of other full curriculums, blog posts, research papers, podcasts, books, it can be as diverse as possible.
Some people can learn best from Articles while others learn best from pure text.
By having a diverse set of learning resources, you’ll learn what works best for you. And in your learning journey. It’s okay to update your curriculum as you go.
Sometimes you might find something new, you’d like to focus on more deeply. And you can allow yourself to do that, As long as you have a set of small achievable learning goals, you’ll have the reward feedback to complete your entire curriculum.
A great easy way to do this is to participate in the 100 days of ML code challenge, It’s a pledge to dedicate at least one hour every day to studying machine learning.
This is a skill that you’ll always be improving at, there’s a bunch of things you could be doing at any given moment, But it’s important to focus on the most important tasks.
multitasking is a myth, switching between tasks is what’s actually happening, And this takes more time, and it’s less productive. Also, it causes more errors.
Every time you switch tasks, you’ve got to regain focus and flow, which takes between five and 20 minutes. I create a to do list of tasks for each day, week, month and year and try to stick to it as best as I can.
It’s worked pretty well so far, evaluate each item on your list as to its importance for your learning goals, and eliminate the unnecessary,
5. Be an active learner
We can use the fast framework to do this effectively. It’s an acronym that stands for forget, activate state teach, and will help you learn anything faster.
Forget means that when you’re learning something new, try and temporarily forget three things. First, forget what you already know about the subject, Lots of people don’t learn faster because they feel like they already know the information.
But our minds are like parachutes, They only work when opened Mic drop. It’s like that Yoda saying you must unlearn what you have learned.
Also, forget about anything that’s not urgent, Remember, the brain can’t multitask, And if you aren’t fully present, you won’t learn as well. And forget your limitations those beliefs that your memory is bad or you don’t have the right education to learn.
The A stands for active most of us were taught in school that learning means being lectured to the opposite is actually true.
We learn by creating information and being active in the process. It’s not a spectator sport. That means asking questions on Stack Overflow.
Taking notes while you consume educational content, and coding up solutions yourself, spend most of your time doing not consuming.
The S stands for state, all learning is state dependent. If you feel bored, you won’t be able to learn as well. So choose states of curiosity, happiness, and eagerness.
You can do this by altering your posture, or ideating about how learning AI will help you in the long run by giving you limitless career options.
T stands for teach. You can literally cut your learning curve in half. If you learn with the intention of teaching it to somebody else. This is what I do full time I learned to teach.
And that’s why I can learn so much so fast, I’m always teaching and while learning start with the simplest explanations first, that way you don’t give up.
For example, if you’re trying to learn about variational auto encoders start with a Article or blog post that explains it at a high level, not the paper.
And as you familiarize yourself with the concept, you can progressively increase the complexity of the educational content you consume.
Oh, and one more thing while you learn Make sure to take frequent breaks, your attention and retention diminish after about 30 minutes so breaks can help reset that.