AI 萌芽

Understand the training loop, loss functions, overfitting, and how to know when your AI model is ready.

Explore how bias enters AI systems, the ethical challenges AI creates, and how we can build fairer technology.

Understand how neural networks learn by propagating errors backwards through layers, using the chain rule to update every weight.

Discover how loss functions measure a model's errors and how optimisers use gradients to systematically reduce them.

Learn how language models break text into tokens using BPE and other algorithms, and why tokenisation shapes everything from cost to capability.

Explore how AI represents words and sentences as vectors in high-dimensional space, enabling semantic search, recommendations, and RAG.

Learn why accuracy alone is misleading, and master the metrics - precision, recall, F1, ROC-AUC, BLEU, and perplexity - that truly measure AI performance.

How GPT, Claude and other LLMs work under the hood

Understand the two most common machine learning failure modes — overfitting and underfitting — with clear examples and how to fix them.

Learn how feature engineering transforms raw data into powerful machine learning inputs — the skill that separates good models from great ones.

A clear comparison of supervised and unsupervised machine learning — when to use each approach, with real-world examples and algorithms.

Learn how decision trees work, why they're one of the most intuitive ML algorithms, and when to use them.

Understand clustering — a key unsupervised learning technique — through K-Means, hierarchical clustering, and real-world applications.