Part III dives into the core of artificial intelligence: the algorithms and architectures that enable machines to learn from data. We progress from classical machine learning foundations to modern deep learning systems.
These chapters connect the biological learning principles from Part I with the computational frameworks that have revolutionized AI. You’ll understand not just how these algorithms work, but why they’re designed the way they are, and how neuroscience continues to inform their development.
Key Themes
Classical ML: Regression, classification, and clustering fundamentals
Deep Neural Networks: Architectures inspired by biological vision
Backpropagation: The algorithm that enabled deep learning
Optimization: Gradient descent and its variants
Regularization: Preventing overfitting and improving generalization
Bio-Plausibility: How artificial learning relates to biological learning
Chapters in This Part
Chapter 12: The Building Blocks: Classical Machine Learning Foundations Linear models, SVMs, decision trees, and clustering algorithms
Chapter 13: Deep Learning: Training & Optimisation Neural network architectures, backpropagation, and modern training techniques
What You’ll Learn
By the end of Part III, you will understand:
- ✓ Fundamental machine learning algorithms and when to use them
- ✓ The mathematical principles underlying neural networks
- ✓ How backpropagation enables efficient learning
- ✓ Modern optimization techniques (Adam, batch normalization, dropout)
- ✓ Convolutional and recurrent network architectures
- ✓ The connections and differences between artificial and biological learning
- ✓ Practical considerations for training deep networks
Deep learning has transformed AI by learning hierarchical representations, a principle borrowed from how brains process information.