Lesson Plan

• Describe the structure of an artificial neuron and its activation function.
• Explain how back‑propagation updates network weights to minimise loss.
• Analyse how ANNs enable automatic feature extraction and non‑linear modelling.
• Evaluate real‑world applications of different ANN types (CNN, RNN, Transformer).
• Compare limitations of ANNs such as data requirements and interpretability.

• Projector or interactive whiteboard
• Slides with neural‑network diagrams
• Sample Python notebook (Jupyter) demonstrating a simple MLP on MNIST
• Handout summarising activation functions and gradient update formula
• Worksheets for group analysis of application case studies
• Computers with internet access (GPU‑enabled environment optional)

Begin with a quick poll: Who has used AI‑powered tools like ChatGPT or image‑recognition apps? Briefly recap that these systems rely on artificial neural networks, linking to prior knowledge of functions and algorithms. Explain that today students will identify how ANNs have advanced machine learning and will be able to articulate their core mechanisms and real‑world impacts.

1. Do‑Now (5') – Students write down one everyday technology powered by AI and share; teacher notes common themes.
2. Mini‑lecture (10') – Present neuron model, activation functions, and back‑propagation using slides and whiteboard.
3. Guided demo (15') – Walk through a Jupyter notebook training a simple MLP on MNIST, highlighting loss reduction and weight updates.
4. Group activity (12') – Teams analyse case studies (CNN for image detection, RNN for speech, Transformer for language) and fill a comparison chart.
5. Whole‑class discussion (8') – Discuss scalability, transfer learning, and limitations; teacher checks understanding with concept questions.
6. Exit ticket (5') – Students write one way ANNs have changed machine learning and one limitation they must consider.

Summarise that ANNs provide flexible, non‑linear models that learn features automatically, enabling breakthroughs in vision, speech, and language. Collect exit tickets, remind students to review the notebook for homework, and ask them to modify the network architecture to observe performance changes.