ML4EA

Learn

Move from foundations to engineering decisions.

Use the chapters in sequence or enter through the mathematical and computational background most relevant to your work.

Choose a starting point

Four useful entry paths

Refresh the mathematicsLinear algebra, probability, statistics, and optimization.Start machine learningLearning types, generalization, evaluation, and data handling.Learn through methodsRegression, classification, ensembles, neural networks, and beyond.Think at system levelValidation, reliability, deployment, autonomy, and governance.

Part I · Chapters 1–4

Mathematical foundations

Build or refresh the concepts that make later algorithms interpretable.

  1. 01

    Introduction to AI and Machine Learning in Engineering

    Places AI and machine learning within engineering practice, contrasting data-driven and model-driven approaches while establishing the book's learning design.

    • AI and machine learning
    • History of AI
    • Modern engineering practice
  2. 02

    Linear Algebra Essentials

    Develops the vector, matrix, geometric, and decomposition tools used to represent engineering data and understand machine learning algorithms.

    • Vectors and matrices
    • Vector spaces and basis
    • Norms and distance
  3. 03

    Probability and Statistics Fundamentals

    Builds the probabilistic language needed to describe uncertainty, summarize samples, perform inference, and evaluate relationships in engineering data.

    • Sampling and descriptive statistics
    • Probability
    • Random variables and distributions
  4. 04

    Optimization Basics

    Introduces objectives, constraints, convexity, and numerical algorithms as the mathematical engine behind model training and engineering design decisions.

    • Objectives and design variables
    • Unconstrained optimization
    • Constrained optimization

Part II · Chapters 5–15

Concepts and algorithms

Study individual learning paradigms, then connect them into complete engineering systems.

  1. 05

    Introduction to Machine Learning

    Establishes the core machine learning vocabulary, learning paradigms, generalization concepts, evaluation practices, and practical Python environment.

    • Types of machine learning
    • Features, labels, and models
    • Empirical risk and generalization
  2. 06

    Supervised Learning: Regression

    Develops regression from linear models through nonlinear basis functions and regularization, with attention to diagnostics and engineering interpretation.

    • Linear regression
    • Parameter estimation
    • Model diagnostics
  3. 07

    Supervised Learning: Classification

    Frames categorical prediction and compares interpretable, geometric, and instance-based classifiers for engineering decisions.

    • Classification formulation
    • Logistic regression
    • Decision trees
  4. 08

    Ensemble Methods

    Explains how bagging, random forests, and boosting combine models to improve predictive performance, robustness, and interpretability.

    • Bias and variance
    • Bagging
    • Random forests
  5. 09

    Neural Networks and Deep Learning

    Moves from artificial neurons and multilayer perceptrons to modern deep architectures, training practices, evaluation, and engineering deployment.

    • Perceptrons and MLPs
    • Activation functions
    • Backpropagation
  6. 10

    Unsupervised Learning

    Introduces methods for discovering structure without labels, including clustering, dimensionality reduction, density estimation, and anomaly detection.

    • K-means and hierarchical clustering
    • DBSCAN
    • PCA
  7. 11

    Reinforcement Learning

    Develops sequential decision-making from Markov decision processes through value, policy, actor-critic, and deep reinforcement learning methods.

    • Markov decision processes
    • Value functions
    • Q-learning and SARSA
  8. 12

    Generative Models

    Surveys probabilistic and neural approaches that generate data, designs, and engineering assistance, including VAEs, GANs, diffusion models, and LLMs.

    • Generative modeling concepts
    • Variational autoencoders
    • Generative adversarial networks
  9. 13

    Physics-Informed Machine Learning

    Combines governing equations, constraints, data, and learning algorithms to build models that respect engineering knowledge and limited-data conditions.

    • Governing equations and constraints
    • Physics-augmented losses
    • Physics-informed neural networks
  10. 14

    Specialized ML Techniques and Emergent Topics

    Connects a broad set of specialized methods to engineering practice, from transfer and Bayesian learning to graphs, symbolic methods, agentic AI, and MLOps.

    • Transfer and Bayesian learning
    • Recommendation and association rules
    • Graph learning
  11. 15

    Integrating Machine Learning into Engineering Systems

    Brings the methods together at system level, emphasizing paradigm selection, data-versus-physics tradeoffs, validation, deployment, trust, and human oversight.

    • Algorithms to engineering decisions
    • Paradigm selection
    • Data-versus-physics design matrix

Learn through application

Run the engineering examples.

Explore Application Examples