Computing and Networks
Machine Learning: A Field Guide
The complete student field guide
AI generated by Nikhil Kumar Rajput
In the reading list AI and machine learning (book 1 of 4)
8 chapters
Contents¶
Front matter
Chapters
Chapter 1: Machine Learning Foundations
- 1.1 What Machine Learning Is: Learning From Data Instead of Writing Rules
- 1.2 AI vs Machine Learning vs Deep Learning vs Generative AI
- 1.3 Supervised and Unsupervised Learning
- 1.4 Self-Supervised, Semi-Supervised and Reinforcement Learning
- 1.5 Key Vocabulary: Features, Labels, Models, Training and Inference
- 1.6 The End-to-End Machine Learning Workflow
- 1.7 Maths Gently I: Vectors, Matrices and Tensors
- 1.8 Maths Gently II: Derivatives, Gradients and Gradient Descent
- 1.9 Maths Gently III: Probability, Distributions and Bayes
- 1.10 Maths Gently IV: Statistics Every ML Practitioner Needs
- 1.11 A Short History of Machine Learning: 1943 to 2025
- 1.12 Machine Learning in Everyday Life
- 1.13 When Not to Use Machine Learning
- 1.14 The Machine Learning Landscape in 2026
Chapter 2: Data, Features & Evaluation Setup
- 2.1 Where Data Comes From: Sources, Formats and Collection
- 2.2 Labelling and Annotation: Getting the Right Answers
- 2.3 Weak Supervision, Active Learning and Synthetic Data
- 2.4 Exploratory Data Analysis and Visualisation
- 2.5 Data Cleaning: Missing Values, Outliers and Duplicates
- 2.6 Feature Engineering I: Scaling, Transforming and Combining Numbers
- 2.7 Feature Engineering II: Categories, Text and Dates
- 2.8 Feature Selection: Keeping the Inputs That Matter
- 2.9 Train, Validation and Test Splits, Cross-Validation and Time-Series Splits
- 2.10 Data Leakage: When Your Model Cheats Without You Noticing
- 2.11 Imbalanced Data: Rare Events and Skewed Classes
- 2.12 Data Augmentation: Making More From What You Have
- 2.13 Famous Datasets Every Student Meets
- 2.14 Data Quality, Versioning and Documentation
Chapter 3: Classical Machine Learning Algorithms
- 3.1 Linear Regression: Least Squares and Its Assumptions
- 3.2 Logistic Regression: Probabilities for Yes/No Questions
- 3.3 Regularised Linear Models: Ridge, Lasso and Elastic Net
- 3.4 k-Nearest Neighbours and Distance-Based Learning
- 3.5 Support Vector Machines and Kernels
- 3.6 Decision Trees: Gini, Entropy and Pruning
- 3.7 Tree Ensembles: Random Forests and Gradient Boosting
- 3.8 Clustering: k-Means, Hierarchical, DBSCAN/HDBSCAN and Gaussian Mixtures
- 3.9 Dimensionality Reduction: PCA, t-SNE and UMAP
- 3.10 Anomaly Detection: Isolation Forests, One-Class SVMs and More
- 3.11 Recommender Systems: Collaborative Filtering and Matrix Factorisation
- 3.12 Time-Series Forecasting: ARIMA, Prophet, Boosting and Foundation Models
- 3.13 Naive Bayes and Bayesian Methods
- 3.14 Choosing an Algorithm: A Practical Decision Guide
Chapter 4: Training, Optimisation & Evaluation
- 4.1 Loss Functions: What the Model Actually Minimises
- 4.2 Gradient Descent in Practice: Batch, Mini-Batch and SGD
- 4.3 Optimisers: Momentum, RMSProp, Adam, AdamW, Lion and Muon
- 4.4 Learning-Rate Schedules and Warm-Up
- 4.5 Bias, Variance, Overfitting and Double Descent
- 4.6 Regularisation During Training: Early Stopping, Dropout and Weight Decay
- 4.7 Classification Metrics: Confusion Matrix, Precision, Recall, F1 and AUC
- 4.8 Probability Calibration and Decision Thresholds
- 4.9 Regression Metrics: MAE, RMSE, R² and Friends
- 4.10 Ranking and Generation Metrics: NDCG, BLEU, ROUGE and Perplexity
- 4.11 Hyperparameter Tuning: Grid, Random, Bayesian and Optuna
- 4.12 Model Selection and Statistical Significance
- 4.13 Interpretability: Feature Importance, SHAP, LIME and Partial Dependence
- 4.14 Experiment Tracking and Reproducible Training Runs
Chapter 5: Deep Learning
- 5.1 Neurons, Layers, MLPs & Activation Functions
- 5.2 Backpropagation & Automatic Differentiation
- 5.3 Initialisation, Normalisation & Residual Connections
- 5.4 Convolutional Neural Networks (CNNs)
- 5.5 Recurrent Networks: RNNs, LSTMs & GRUs
- 5.6 Attention & the Transformer Architecture
- 5.7 Embeddings & Representation Learning
- 5.8 Transfer Learning & Fine-Tuning
- 5.9 Autoencoders, VAEs & GANs
- 5.10 Diffusion Models & Flow Matching
- 5.11 Vision Transformers & CLIP
- 5.12 Graph Neural Networks
- 5.13 State-Space Models, Hybrids & Mixture-of-Experts
- 5.14 Training at Scale: Precision & Parallelism
Chapter 6: LLMs, Generative AI & Reinforcement Learning
- 6.1 Language Modelling & Tokenisation
- 6.2 Pretraining & Scaling Laws
- 6.3 Instruction Tuning & Alignment: SFT, RLHF, DPO, RLVR
- 6.4 Reasoning Models & Test-Time Compute
- 6.5 Prompting & Context Engineering
- 6.6 Retrieval-Augmented Generation (RAG)
- 6.7 Agents, Tool Use & MCP
- 6.8 Multimodal Models & Generative Media
- 6.9 Parameter-Efficient Fine-Tuning (LoRA, QLoRA) & Distillation
- 6.10 Quantisation & Efficient Inference
- 6.11 Evaluating LLMs: Benchmarks, Arenas & Contamination
- 6.12 The 2026 Model Landscape: Frontier & Open-Weight Families
- 6.13 Small & On-Device Models
- 6.14 Reinforcement Learning Fundamentals
Chapter 7: Tools, Frameworks & MLOps
- 7.1 The Python ML Stack
- 7.2 Deep-Learning Frameworks: PyTorch, JAX and TensorFlow/Keras
- 7.3 The Hugging Face Ecosystem
- 7.4 The MLOps Lifecycle and Maturity Levels
- 7.5 Model Registries, Artifacts and Lineage
- 7.6 Feature Stores
- 7.7 Pipelines and Orchestration
- 7.8 Model Serving and Deployment
- 7.9 Monitoring, Data Drift and Concept Drift
- 7.10 CI/CD and Testing for Machine Learning
- 7.11 Cloud ML Platforms
- 7.12 Hardware for ML: GPUs, TPUs, NPUs and Memory Maths
- 7.13 Cost Management for ML and AI
- 7.14 LLMOps: Prompts, Evaluations, Guardrails and Observability
Chapter 8: Responsible AI, Careers & Learning
- 8.1 Bias and Fairness
- 8.2 Privacy: Differential Privacy, Federated Learning and Anonymisation
- 8.3 Security of ML and LLM Systems
- 8.4 Explainability, Transparency and Documentation
- 8.5 AI Safety and Alignment Basics
- 8.6 The EU AI Act: Risk Tiers, Obligations and Timeline
- 8.7 AI Standards and Policy Around the World
- 8.8 The Environmental Cost of AI
- 8.9 Careers in Machine Learning and AI
- 8.10 Certifications
- 8.11 Courses, Books and Free Learning Resources
- 8.12 Projects, Kaggle and Building a Portfolio
- 8.13 Interview Questions with Model Answers
- 8.14 A 16-Week Study Plan
Back matter
Ten Things Every Machine Learning Student Should Master
- 1. Start with the problem, not the algorithm
- 2. Data quality sets the ceiling
- 3. Evaluate honestly, without leakage
- 4. Balance bias and variance
- 5. Pick the right tool for the data
- 6. Understand transformers and LLMs from the inside
- 7. Deployment is the beginning, not the end
- 8. Do the memory and cost maths
- 9. Build responsibly — and lawfully
- 10. Prove your skills and keep learning
About this book¶
A student's plain-English guide to machine learning — data, classic algorithms, deep learning, transformers and LLMs, MLOps, responsible AI and careers. Every topic is explained with an everyday analogy, a worked example, step-by-step mechanics, a diagram, and a table of the algorithms, models, techniques, tools & concepts you need to know.