Basic Information
| Field | Details |
|---|---|
| English name | AI / ML Core Concepts |
| Full name | Artificial Intelligence / Machine Learning / Deep Learning / Training / Inference |
| Chinese description | 人工智能与机器学习核心概念 |
| Japanese description | AI・機械学習の基本概念 |
| Exam frequency | ⭐⭐⭐⭐⭐ |
| Often confused with | AI / ML / Deep Learning / Generative AI / Training / Inference |
In one sentence
First separate AI, ML, deep learning, and generative AI, then decide whether the task is training or inference.
Key points
- AI is the broad field, ML is a subset of AI, and deep learning is a subset of ML.
- Training learns parameters from historical data; inference applies the model to new input.
- Data quality, schema, consistency, and freshness often matter more than blindly increasing model complexity.
- Pretrained AI APIs, custom ML, foundation models, and ready-made assistants are different abstraction levels.
Exam focus
- Start with input, output, need for training, customization level, and operational burden.
Common pitfalls
- Do not choose by product name alone; confirm data type, latency, control, operations, and cost constraints.
Remember
Decide whether to use a ready capability, train a model, or call a foundation model.