Non-Medium Members Can Read the Full Article by Clicking Here.
👆 👆👆👆👆
Welcome to my fourth article in this series!
As a Psychologist, I've been always interested in how people study and learn.
So in this series, I aim to explain in detail all the principles of ultralearning proposed by Scott Young in his book "Ultralearning"!
I shared with you how anyone can learn any topic with the power of ultralearning (learn intensively about any topic).
So far, our list of principles is as follows:
Today we will learn about N4: Drill and attack your weakest point.
Learning as a Chemistry Reaction
In chemistry, the rate-determining step is the slower part of a chain of reactions that will define the amount of time needed for the entire reaction to occur.
Scott argues that learning works similarly.
So by identifying a rate-determining step in the learning reaction, we can isolate it and work on it specifically.
This is the strategy behind drills.
When learning, some concepts will determine how much time we will master a topic.
For example, artificial intelligence (AI) requires a lot of math.
I'm a psychologist with no background in math, so that took me a considerable amount of time.
However, I drilled all the math required for AI to master it faster.
In short, drill refers to learning your weakest point in your ultra-learning project.
The Direct-Then-Drill Approach
Drills require a few steps:
- Practice the skill directly:
Figure out where and how the skill will be used and try to match that situation when practicing.
For example, learn programming by writing software, or practice a language by speaking it.
2. Analyze the skill and isolate components:
Find components that are rate-determining or subskills that are hard to master (like math in AI in my previous example).
Then practice them separately.
3. Go back to direct practice and integrate what you've learned:
This will give you an intuition of how well your drill was designed.
Additionally, this will help you to recap everything you drilled and learned, and how it can be adapted to your learning context.
When I was organizing my ultra-learning project to learn AI, I saw that I needed to have a good understanding of math.
Particularly, linear algebra, calculus, statistics, and probability.
I'm a psychologist, and I never studied those topics, except statistics.
So in my case, my drill was focused on math.
- First, I isolated the problem into brief study sessions, and tried to align what I was learning with what I was programming.
- Second, instead of using popular Python modules, I tried to do vectorization and calculus using only numpy.
- Third, I recap everything and see my improvements.
This is how I drilled math to understand how machine learning works.
3 Problems When Designing Drills
When designing your drills, you will probably face 3 problems:
- To figure out when and what to drill.
- Designing a drill for improvement.
- Drills are hard and uncomfortable.
This leads us to Ultralearning Principle N1: Meta-Learning.
Focus on the skills you find difficult and are fundamental for learning the topic.
For example, if you're learning Neuroscience with a background in engineering, you will need to drill in Neurobiology.
Always drill in what makes you feel uncomfortable.
Don't feel bad if a topic is too hard.
Everyone can learn anything. Don't give up!
Drill!
Conclusion
That's it for the principle N4.
When learning a new topic we will face uncomfortable topics that will be hard to master.
In this article, we learned how to drill to face those problems.
In the next one, we will learn how to test your knowledge effectively with the retrieval principle.
🤓R I#39m writing an article summarizing each of the 9 principles, so follow me, subscribe, and stay tuned.
See you in the next article!
🦜 If you liked my post, please clap 👏 50 times for more content about Neuroscience, Psychology, Productivity, Animals, and Python!
And don't forget to follow me for more!
Thanks for reading!