This week we learned the most fundamental concepts of AI and machine learning step by step. Today we will weave together the content from Day 1 through Day 4 for review, and go over the parts most often confused on the exam once more.
Review is not mere memorization — it is a time to redraw "how the concepts connect." This week's core flow is as follows.
[Day1] The AI > ML > DL relationship
↓
[Day2] ML learning types (supervised, unsupervised, reinforcement)
↓
[Day3] Problems where ML fits vs. where it doesn't
↓
[Day4] Key terms (model, training, inference, feature, label, overfitting)
The first thing to remember is the nested relationship.
All deep learning is machine learning, but not all AI is machine learning. A system whose rules were written directly by a human may be AI, but it is not machine learning.
Machine learning is divided into three types by learning approach.
| Learning type | Answers (labels) | Representative tasks |
|---|---|---|
| Supervised learning | Present | Classification (categories), regression (numbers) |
| Unsupervised learning | Absent | Clustering, dimensionality reduction |
| Reinforcement learning | Replaced by rewards | Games, robotics, control |
Let's revisit the two most commonly confused pairs.
Machine learning is not a cure-all. Judge its suitability by the following criteria.
Click a choice to reveal the answer and explanation.
Question 1
Which statement about the relationship between AI, machine learning, and deep learning and about learning types is correct?
Question 2
"Automatically grouping customer data into similar tendencies without answers" and "predicting tomorrow's temperature as a number" are, respectively, which tasks?
Question 3
In a certain business task, the rules are very simple and clear ("amount × fixed tax rate"), and the results must always be 100% accurate. Which judgment is most appropriate?
Question 4
A trained model is nearly perfect on the training data but its performance drops sharply on new data. You are also asked for the definitions of feature and label. Which statement is correct?
Question 5
Putting together what was learned this week, which is the most appropriate judgment for using machine learning well?
data + rules → resultsdata + results → rules| Machine learning fits | Machine learning doesn't fit |
|---|---|
| Rules are complex or hard to define | Rules are simple and clear |
| Data is plentiful | There is almost no data |
| Some margin of error is acceptable | 100% accuracy and full explainability are mandatory |
Remember that choosing the simplest method that is sufficient is good design.
Let's redraw the machine learning flow in terms of the vocabulary.
Features + labels [training data]
↓ Training
[Model] completed
↓ Inference
New input (features) → predicted label
↓
Evaluate with [test data] → check for overfitting/underfitting
💡 Related theory: This week's content is the foundation of the "AI/ML fundamental concepts" domain of the AIF-C01 exam. You need a clear grasp of this flow to comfortably follow the advanced topics ahead, such as generative AI and AWS AI services.
Here are the points where mistakes are easy to make on the exam.
| Trap | Correct understanding |
|---|---|
| "Deep learning is the broadest concept" | AI is the broadest. Deep learning is the narrowest |
| "Classification and regression are the same" | Classification = categories, regression = numbers |
| "Classification and clustering are the same" | They differ in the presence of answer labels |
| "Machine learning is always better" | The traditional approach wins for simple problems |
| "It's fine to evaluate with training data" | Honest evaluation requires test data |
| "Overfitting = not enough learning" | Overfitting is actually memorizing too much |
Compressing this week into one sentence per day:
Starting next week, we go one step further on the foundation learned today. We will expand into how machine learning and deep learning are actually used, and into the concept of generative AI, which is drawing the most attention these days. Be sure to consolidate this week's core flow.