Assitan Koné
Jun 30

# How I Leveraged the Feynman Technique to Teach Myself Data Science

Learning data science can be a daunting task and very overwhelming if we don't have a plan.

One day I heard about the Feynman Technique, named after the renowned physicist Richard Feynman.
It involves breaking down complex topics into comprehensible explanations that even a beginner can grasp.

By teaching these simplified explanations to someone else (or an imaginary student), you reinforce your own understanding and identify gaps in your knowledge.

This iterative process of simplification and refinement forms the cornerstone of effective learning.

## Watch the video:

#### Author

Assitan Koné
Founder @Codistwa
Empty space, drag to resize

## Step 1: I Learned the basics intuitively

Data science is built upon a foundation of mathematics, statistics, and programming.

First, I focused on intuitively grasping fundamental concepts before diving into technical details. This is important to avoid being overwhelmed. For instance, I tackled probability by imagining a simple coin-flipping scenario, connecting abstract ideas to the real world.
This is perfect because we can do that in real life. Then, I wrote down these concrete examples to know to forget the context of these complex concepts. You can write, and even draw! I prefer the second technique to teach.

Rolling dice is also a great way to understand probability fundamentals

## Step 2: I taught as if explaining to a novice

Once I felt comfortable with my intuitive understanding, I imagined myself explaining data science concepts to a complete novice.
I chose a friend with limited technical knowledge as my imaginary student. But it also can be an actual friend or even better, your mom! Usually, our parents don't understand very well our tech job. Or if you are a parent yourself and you have a child that I have at least 5 years old, you can try to teach him/her the concept.

## Step 3: I identified gaps and refine

While teaching, I discovered knowledge gaps that acted as signposts for further exploration. For example, while explaining regression analysis to my imaginary student, I used mind maps to visualize and understand the significance of the slope and intercept.

What I like about mind map is that it's a great way to organize information and it's therapeutic for me.

Find a way to teach so you can remember easily the concepts. So it can be a mind map or just a drawing on a whiteboard, on paper, etc.

## Step 4: I simplified further and repeat

With newfound insights, I refined my explanations even further.
Through an iterative process, I refined my explanations of machine learning algorithms by finding relatable metaphors and analogies, making the topic easier to understand.
Two of my favorite analogies are: recipe instructions, where the ingredients are data, and the output is a delicious prediction and the real estate agent who can predict the price of a house. You can practice this technique with our membership.

## Step 5: I did practical application and practice

No learning journey is complete without practical application. With my simplified explanations, I tackled data science projects and challenges.
The Feynman Technique gave me a solid foundation, enabling me to break down complex problems, select appropriate methods, and effectively communicate my findings. And even if you do classic projects, break down every step prevent being overwhelmed.

This technique gives you a 360 view of complex concepts.

## The rewards of the Feynman Technique in Data Science learning

By using the Feynman Technique, I learned data science and gained an appreciation for its interdisciplinary nature. Breaking down complex concepts into simple explanations was helpful for communicating with peers and colleagues.
Through this technique, I developed a growth mindset, embracing challenges and explaining complex concepts. Eventually, I transformed into a proficient data scientist capable of demystifying the field for both technical and non-technical audiences.

## Conclusion

The Feynman Technique is a powerful tool for mastering complex subjects like data science. By embracing the principles of simplicity, active recall, and iterative refinement, I not only taught myself data science but also developed a lifelong approach to learning that continues to enrich my intellectual pursuits.
So, whether you're a data enthusiast or embarking on any learning endeavor, consider harnessing the power of the Feynman Technique to unlock your full learning potential.

# Machine Learning & Deep Learning Mastery Kit

## Math for Machine Learning Comprehensive Guide

Understand quickly why math is important to do machine learning.