aimnova.
DashboardMy LearningPaper MasteryStudy Plan

Aimnova site navigation

Stay in the loop

Get the latest study resources and updates

New features, study tips and exam insights — straight to your inbox.

IB Diploma

  • IB Past Papers
  • IB Study Notes
  • IB Question Bank
  • IB Mock Exams
  • IB Revision

IB Subjects

  • IB Math AA
  • IB Math AI
  • IB Economics
  • IB Business Management
  • IB Physics
  • IB Biology
  • View all IB subjects→

IB Past Papers

  • IB Math AA HL Past Papers
  • IB Math AA SL Past Papers
  • IB Math AI HL Past Papers
  • IB Math AI SL Past Papers
  • IB Economics HL Past Papers
  • IB Economics SL Past Papers
  • IB ESS Past Papers
  • View all past papers→

Study Resources

  • Study Notes
  • Question Bank
  • Mock Exams
  • Flashcards
  • Revision Guide
  • Exam Skills
  • Command Terms
  • Grade Calculator
  • Exam Timetable 2026

Aimnova

  • Features
  • Pricing
  • For Schools
  • For Parents
  • About Us
  • Blog
  • Contact
aimnova.

AI-powered study platform for smarter revision, past-paper analysis and examiner-style feedback.

TermsPrivacyCookies·© 2026 Aimnova. All rights reserved.8afc4e3

Aimnova is not affiliated with or endorsed by the International Baccalaureate Organization (IB).

NotesComputer Science HLTopic 4.3Neural networks
Back to Computer Science HL Topics
4.3.84 min read

Neural networks (Computer Science HL)

IB Computer Science • Unit 4

Your first topic is free to keep

Know exactly what to write for full marks

Practice with exam questions and get AI feedback that shows you the perfect answer — what examiners want to see.

Start Free

Contents

  • Layers of weighted sums
  • The three kinds of layer
  • How it learns
  • Exam-style question
The big idea: An artificial neural network is layers of simple units. Each unit takes its inputs, multiplies each by a weight, adds them up, adds a bias, and passes the result through an activation function.

Nothing in one unit is complicated. The capability comes from having many, in layers.

Step through: one unit's arithmetic, the layers, what a hidden layer adds, and how training adjusts the weights.

Interactive diagram

Explore the labelled diagram, charts and maps for this topic in full study mode.

Claim your free topic

Free preview

This is the free notes preview

You're reading the free notes. Aimnova Pro unlocks the full study experience — and you can try it with your first topic free to keep:

  • FlashcardsLock in vocabulary and key terms with spaced repetition.
  • Practice questionsAnswer exam-style questions and get instant AI marking.
  • Mock exams & past-paper vaultSit full mocks and see exactly how examiners award marks.
  • Personalised study planA daily plan built around your exam date and weak areas.
Start Studying Free Full access to Aimnova Pro · cancel anytime

Input layer

  • One unit per feature
  • Does no computation — it holds the values
  • 28×28 pixels means 784 input units

Hidden layers

  • Where the work happens
  • Each layer builds on the patterns the previous one found
  • "Deep learning" means several of them

Output layer

  • One unit per class, or one for a regression
  • Its activation turns the numbers into an answer
  • Ten units for the digits 0–9
Why the activation function is not optional: Without it, every layer is a weighted sum of weighted sums — which collapses to one weighted sum. A hundred layers would have exactly the power of one.

The non-linear activation is what lets extra layers add anything at all.

Know your predicted grade

Take timed mock exams and get detailed feedback on every answer. See exactly where you're losing marks.

Try Mock Exams FreeYour first topic is free to keep • No credit card required
1

Forward pass

2

Loss

3

Backpropagation

4

Update

Training adjusts the weights, and nothing else: The structure — how many layers, how many units — is fixed before training and never changes. Those are hyperparameters.

What training produces is the set of weights and biases. That is the whole learned model.
It cannot explain itself: A trained network is millions of numbers. There is no rule to read out, no path to trace, no reason it can give.

That is why an accurate network is often rejected for lending or medicine, where a decision must be justified to the person it was made about.

How this is tested — you must describe the structure layer by layer and explain what training actually changes. It comes up two ways:

Paper 1 Section A

  • Outline the structure and function of an ANN, 3-5 marks
  • Name the three kinds of layer
  • State what a weight does

Paper 1 Section B — case study

  • Explain why multi-layer networks are more capable
  • Judge an ANN against an interpretable model
The classic trap: Saying a network "works like the brain". It is an analogy in name only — a unit is a weighted sum and an activation. Describe the arithmetic, not the metaphor.
IB-style questionOutline[5 marks]

Outline the structure and function of an artificial neural network used to recognise handwritten digits, and explain why a hidden layer is necessary.

Model answer plan

See the mark-by-mark plan — for / against / judgement, with marking guidance — in study mode.

Claim your free topic

IB Exam Questions on Neural networks

Practice with IB-style questions filtered to Topic 4.3.8. Get instant AI feedback on every answer.

Practice Topic 4.3.8 QuestionsBrowse All Computer Science HL Topics

How Neural networks Appears in IB Exams

Examiners use specific command terms when asking about this topic. Here's what to expect:

Define

Give the precise meaning of key terms related to Neural networks.

AO1
Describe

Give a detailed account of processes or features in Neural networks.

AO2
Explain

Give reasons WHY — cause and effect within Neural networks.

AO3
Evaluate

Weigh strengths AND limitations of approaches in Neural networks.

AO3
Discuss

Present arguments FOR and AGAINST with a balanced conclusion.

AO3

See the full IB Command Terms guide →

Related Computer Science HL Topics

Continue learning with these related topics from the same unit:

4.1.1Types of machine learning
4.1.2Hardware for machine learning
4.2.1Data cleaning
4.2.2Feature selection
View all Computer Science HL topics

Improve your exam technique

Command terms, paper structure, and mark-scheme tips for Computer Science HL

Previous
4.3.7Genetic algorithms
Next
CNNs4.3.9

Make these notes count

Reading notes is just the start. Test yourself with IB-style questions and get feedback that shows you what examiners want.

Start FreeView All Computer Science HL Topics