Finding time to take courses in our chosen fields or careers can be challenging, especially with everything going on in our lives — work, social events, family gatherings, and other obligations.

The current wave of acceleration in Artificial Intelligence has made completing tasks easier and faster than ever before, helping us become more efficient and productive — often without requiring deep foundational knowledge in a particular industry or topic.

Yet, even though chatbots have helped us do more with less, there’s still no substitute for continuous education, professional development, and upskilling. True credibility in any field still comes from having both industry experience and foundational skills that allow you to lead with confidence and depth.

I recently completed the Advanced Learning Algorithms online course, which focuses on the fundamentals of Deep Learning (Neural Networks) and Advanced Decision Trees such as Random Forests and XGBoost. The course is designed for those interested in Machine Learning Engineering and Artificial Intelligence, and I’m excited to share my three key takeaways from the experience.

1. Neural Networks Are the Foundation for LLMs

Large Language Models (LLMs) like ChatGPT, Gemini, and Claude might seem cutting-edge, but they’re all built on the same foundation: neural networks.

Simple networks — built from neurons, weights, activations, and gradients — form the building blocks for more advanced architectures.

• Forward propagation passes information through layers.
• Backpropagation calculates errors using derivatives.
• Gradient descent updates weights to minimize loss.

These concepts scale massively into transformer architectures, which power today’s LLMs. Understanding the math and structure behind neural networks makes it much easier to grasp how transformers encode language, attention, and context.

In short, if you understand neural networks, you understand the DNA of LLMs. The complexity changes — the principles don’t.

Here is a diagram of what Neural Networks look like:

2. Foundational Math Builds Machine Learning Intuition

It’s easy to rely on libraries like TensorFlow or PyTorch and skip the equations. But the deeper your understanding of the math, the stronger your intuition becomes.

The course reinforced the importance of linear algebra, calculus, and probability in every part of model training. Here’s why they matter:

• Matrix multiplication enables efficient vectorized operations.
• Derivatives and the chain rule drive backpropagation.
• Gradient descent uses partial derivatives to optimize loss functions.

Grasping these fundamentals transforms your debugging and optimization process. You’ll know why your model is overfitting, why your gradient exploded, or why your learning rate is too high.

Understanding these concepts helps someone move from model user to model architect.

3. Decision Trees Are Powerful — But Metrics Like Entropy and Information Gain Are Key

Decision trees, Random Forests, and XGBoost remain some of the most powerful algorithms in applied machine learning for classification and even regression prediction problems for Marketing-Mix Modeling (MMM). They are generally much more accurate than traditional Decision trees. They however still rely on two key metrics for calculation which are foundational.

• Entropy measures disorder or uncertainty in the data.
• Information gain measures how much uncertainty decreases when you split on a given feature.

Understanding these metrics makes tree-based models far more interpretable. Instead of treating Random Forests or XGBoost as black boxes, you start seeing how they make clear, data-driven decisions — minimizing uncertainty with every branch

Final Thoughts Whether you’re learning foundational knowledge to grow in your career or simply pursuing personal development for your own curiosity, finding time for learning and self-growth is essential. In a world that’s constantly evolving — especially with the rise of AI — taking the initiative to strengthen your understanding and build new skills isn’t just a professional advantage; it’s an investment in yourself.