Step by Step
🏙️
AI — the whole City
The broadest possible territory. Just as a city contains many different neighborhoods with very different characters, AI contains many different techniques and approaches, only some of which involve learning from data at all.
Example: a city has a financial district, residential areas, and industrial zones — AI likewise contains rule-based systems, search algorithms, and machine learning, among other approaches.
🏘️
Machine Learning — one Neighborhood inside the city
A specific, well-defined district within the larger city — narrower than the whole city, but still containing many different "houses" (specific techniques) within it.
Example: the ML neighborhood contains many types of houses — decision trees, support vector machines, and deep neural networks are all different houses within the same ML neighborhood.
🏠
Deep Learning — one specific House in that neighborhood
A single, specific house within the ML neighborhood — you cannot have this house without also being inside the neighborhood, and the neighborhood without also being inside the city.
Example: you can visit the ML neighborhood without ever entering the deep learning house (by using a decision tree instead), but you cannot be in the deep learning house without also being in the ML neighborhood and the AI city.
Applied Walkthrough
1
You're asked to picture where a decision tree algorithm fits in the AI/ML/DL hierarchy.
2
A decision tree is a machine learning technique, but it is NOT a deep learning technique — it doesn't use layered neural networks.
3
Using the city metaphor: the decision tree is a different house within the same ML neighborhood as deep learning — both are inside ML, and therefore both are inside the AI city, but they are different houses.
4
This spatial image helps you quickly place ANY new AI term you encounter: ask whether it's a whole new neighborhood, or just another house within the ML neighborhood you already know.
Exam Application
This spatial metaphor is a second, complementary way to lock in the same core hierarchy tested in the "nested circles" lesson (AI ⊃ ML ⊃ Deep Learning). If one visual image doesn't stick for you, having two independent mental pictures — nested circles AND a city map — doubles your chances of instant recall under exam pressure.
⚠ Common Trap
Don't let having two metaphors (nested circles and the city) confuse you into thinking there are two different hierarchies. They describe the exact same relationship — AI is broadest, ML is the middle layer, DL is the narrowest and most specific — just visualized two different ways. Use whichever image locks in faster for you personally.
✓ Quick Self-Check
1. In the city metaphor, what does the whole city represent?
Artificial Intelligence — the broadest category.
Tap to reveal / hide
2. In the city metaphor, what does one neighborhood represent?
Machine Learning — a specific district within the broader AI city.
Tap to reveal / hide
3. In the city metaphor, what does one house represent?
Deep Learning — one specific technique within the ML neighborhood.
Tap to reveal / hide
4. Can you be in the deep learning house without also being in the ML neighborhood and the AI city?
No — just like in a real city, you cannot be inside a house without also being inside its neighborhood and its city.
Tap to reveal / hide
5. Does a decision tree algorithm belong in the deep learning house?
No — a decision tree is a different house within the same ML neighborhood; it does not use layered neural networks, so it isn't deep learning.
Tap to reveal / hide