Chapter 7: Machine Learning and Artificial Intelligence · Term 2
Lesson 7-2 explained: Neural Networks and Deep Learning
A neural network = layers of neurons (input → hidden → output); each neuron weights and sums its inputs, and training tunes the weights on data. Deep learning stacks many hidden layers to build features (edges → parts → wholes), enabled by big data and GPUs.
Lesson summary
A neural network is layers of connected neurons: an input layer, hidden layers that extract features, and an output layer. Each neuron multiplies its inputs by weights and sums them; training adjusts those weights from data. Deep learning stacks many hidden layers to build features step by step — the leap behind modern AI.
Takeaway: A neural network = layers of neurons (input → hidden → output); each neuron weights and sums its inputs, and training tunes the weights on data. Deep learning stacks many hidden layers to build features (edges → parts → wholes), enabled by big data and GPUs.
Key terms
- Node (neuron)
- Layer
- Input / hidden / output layer
- Weight
- Training
- Training data
- Generalize
- Deep learning
Textbook page: Arabic book 59 · English book 68
Free questions on this lesson
From the textbook question bank. Think of the answer first, then open it.
1. What is a neural network made of?
- Rules written by hand for every case
- Real brain cells inside the computer
- Many connected nodes (neurons) arranged in layers
- One big table of numbers
Answer
Correct answer: Many connected nodes (neurons) arranged in layers
A neural network is simple processing units (nodes) arranged in layers, with nodes of adjacent layers connected.
1. The Structure of Neural Networks · page 69
2. In the neuron figure, what does a thin light line between an input and the neuron mean?
- A large weight that strongly reflects the input
- That the input is the final answer
- That the input is not connected
- A small weight: the neuron reflects that input only a little
Answer
Correct answer: A small weight: the neuron reflects that input only a little
In the figure's key: thick line = large weight, medium line = medium weight, thin line = small weight.
2. How a Neuron Works · page 70
3. What became possible thanks to deep learning in many tasks once hard for computers?
- Accuracy close to or above human performance on specific tasks
- 100% correct judgments on every problem
- Doing without data entirely
- Full general understanding like a human in everything
Answer
Correct answer: Accuracy close to or above human performance on specific tasks
Deep learning reached high accuracy on specific tasks, but this does not mean general understanding or always being right.
4. Why Deep Learning Brought a Leap in AI · page 71
Keep studying on the platform
The full explanation of every lesson, interactive simulations, homework and an exam per chapter, and the whole question bank. Try it free, no code needed.
Lessons in this chapter 7: Machine Learning and Artificial Intelligence
- 7-1 The Basics of Machine Learning
- 7-2 Neural Networks and Deep Learning
- 7-3 Large Language Models (LLM) and Generative AI