What sets deep learning apart from classic machine learning?
Why A is the answer
Multi-layer networks that learn features from data
Layers learn their own features.
In detail
- Layers learn their own features.
- Deep learning needs more data.
Use case
An illustrative scenario, not a real incident.
A classic model relies on hand-built features, and a deep network learns its own from images.
See it in different ways
flowchart LR A["Artificial intelligence"] --> B["Machine learning"] B --> C["Deep learning"] C --> D["Generative AI"]
mindmap
root((AI))
Machine learning
Deep learning
Generative AI
Drivers
GPUs
Data
Algorithms
Stages
Training
Inferenceflowchart TD
Q{"Which approach?"}
Q -->|"People"| A0["Classic ML"]
Q -->|"The network"| A1["Deep learning"]Key point: Deep networks learn features.
Why the other options are wrong
- B. It needs no data at all during training
- It needs lots of data.
- C. It only runs on the CPU in a single thread
- It benefits from GPUs.
- D. All of its rules are written by hand
- Rules are learned, not written.