Get the app

AI & Machine Learning

A comprehensive, 3-pillar curriculum that builds durable understanding of artificial intelligence and machine learning — from the mathematical foundations and classical algorithms, through modern deep learning architectures, to the societal implications of deploying AI in the real world. Emphasizes timeless principles over transient tools so knowledge remains valuable as the field evolves.

670 lessons · about 1000 hours

Download on the App Store
Explore with AI:ChatGPTClaudePerplexity
Foundations of Intelligent Systems
Deep Learning & Modern Architectures
AI in the Real World

What you'll learn

Foundations of Intelligent Systems

The mathematical and algorithmic bedrock of AI and machine learning. Covers the math you need (linear algebra, calculus, probability, statistics), classical ML algorithms, optimization, and probabilistic reasoning. These principles are timeless — they haven't changed in decades and won't change in the next decade.

  1. Unit 1

    Mathematical Foundations for Machine Learning

    Build the essential mathematical toolkit required for understanding machine learning algorithms. This course covers linear algebra, calculus, and core mathematical concepts that form the backbone of every ML technique, from simple linear regression to deep neural networks.

    22 lessons
    1. Vectors and Scalars
    2. Matrix Operations
    3. Eigenvalues and Eigenvectors
    4. Derivatives and Rates of Change
    5. Inner Products and Their Properties
    6. Dot Product and Projections
    7. Linear Transformations
    8. Eigendecomposition
    9. Partial Derivatives and Gradients
    10. Orthogonality and Orthogonal Projections
    11. Norms and Distance Measures
    12. Determinants and Invertibility
    13. Singular Value Decomposition (SVD)
    14. The Chain Rule
    15. The Gram-Schmidt Process
    16. Linear Independence and Basis
    17. Systems of Linear Equations
    18. Positive Definite and Semi-Definite Matrices
    19. Taylor Series Approximation
    20. From Inner Products to the Kernel Trick
    21. Vector Spaces and Subspaces
    22. Jacobian and Hessian Matrices
  2. Unit 2

    What Is Artificial Intelligence

    A comprehensive introduction to the field of artificial intelligence, tracing its origins from the philosophical questions about machine thought through the landmark achievements that have shaped the modern AI landscape. This course covers the key definitions, paradigms, and debates that every AI practitioner should understand.

    22 lessons
    1. Early Automata and Mechanical Thought
    2. Defining Intelligence
    3. Narrow AI in Practice
    4. Deep Blue and Chess
    5. The Turing Test
    6. Turing's Foundational Contributions
    7. Symbolic AI (GOFAI)
    8. Artificial General Intelligence (AGI)
    9. Watson and Jeopardy!
    10. The Chinese Room Argument
    11. The Dartmouth Conference and the Birth of AI
    12. Statistical and Connectionist AI
    13. Superintelligence and Existential Risk
    14. AlphaGo and the Game of Go
    15. Consciousness and Machines
    16. AI Winters and Hype Cycles
    17. Hybrid and Neurosymbolic Approaches
    18. Measuring Progress Toward AGI
    19. GPT and the Large Language Model Revolution
    20. Ethics and Responsibility in AI
    21. The Modern AI Resurgence
    22. AlphaFold and Scientific Discovery
  3. Unit 3

    Model Evaluation and Selection

    Master the tools and techniques for rigorously evaluating machine learning models and selecting the best model for a given problem. This course covers evaluation metrics, cross-validation strategies, hyperparameter tuning methods, and the statistical considerations needed to draw reliable conclusions about model performance.

    22 lessons
    1. The Confusion Matrix
    2. MSE, RMSE, and MAE
    3. K-Fold Cross-Validation
    4. Parameters vs. Hyperparameters
    5. Statistical Significance of Model Differences
    6. Accuracy and Its Limitations
    7. R-Squared and Adjusted R-Squared
    8. Stratified and Grouped Cross-Validation
    9. Grid Search
    10. Probability Calibration
    11. Precision, Recall, and F1 Score
    12. MAPE and Relative Error Metrics
    13. Time Series Cross-Validation
    14. Random Search
    15. Learning Curves as Diagnostic Tools
    16. ROC Curve and AUC
    17. Residual Analysis
    18. Nested Cross-Validation
    19. Bayesian Optimization
    20. Systematic Model Selection Workflow
    21. Precision-Recall Curve and Average Precision
    22. Early Stopping and Successive Halving
  4. Unit 4

    How Machines Learn

    Understand the fundamental principles behind machine learning at a conceptual level, including the major learning paradigms, the role of loss functions and optimization, and the critical tradeoffs that govern every ML system. This course bridges the gap between knowing what AI is and understanding how it actually works.

    21 lessons
    1. Supervised Learning
    2. Hypothesis Space
    3. Loss Functions
    4. Bias-Variance Decomposition
    5. Regularization Intuition
    6. Unsupervised Learning
    7. Inductive Bias
    8. Empirical Risk Minimization
    9. Overfitting
    10. L1 and L2 Regularization
    11. Reinforcement Learning
    12. Model Complexity
    13. Structural Risk Minimization
    14. Underfitting
    15. Early Stopping and Dropout
    16. Self-Supervised Learning
    17. Representation and Feature Spaces
    18. Surrogate Loss Functions
    19. Learning Curves
    20. The No Free Lunch Theorem
    21. Double Descent and Modern Interpolation
  5. Unit 5

    Supervised Learning

    A comprehensive tour of the most important supervised learning algorithms, from classical linear models to powerful ensemble methods. This course covers the mathematical intuition, practical strengths and weaknesses, and appropriate use cases for each algorithm, preparing you to select and apply the right model for any supervised learning problem.

    22 lessons
    1. Linear Regression
    2. Decision Trees
    3. K-Nearest Neighbors (k-NN)
    4. Maximum Margin Classifier
    5. Why Ensembles Work
    6. Polynomial and Basis Function Expansion
    7. Tree Pruning and Complexity Control
    8. The Curse of Dimensionality
    9. Support Vectors and Soft Margin
    10. Bagging and Pasting
    11. Logistic Regression
    12. Random Forests
    13. Naive Bayes Classifier
    14. Kernel SVM
    15. Boosting Principles: AdaBoost and Beyond
    16. Regularized Linear Models: Ridge, Lasso, and Elastic Net
    17. Gradient Boosting
    18. Choosing k and Distance Metrics
    19. SVM for Regression (SVR)
    20. Stacking (Stacked Generalization)
    21. Multiclass Extensions
    22. Feature Importance in Tree-Based Models
  6. Unit 6

    Probabilistic Graphical Models

    Explore the powerful framework of probabilistic graphical models that combine probability theory with graph theory to represent and reason about complex systems with many interacting variables. This advanced course covers Bayesian networks, Markov random fields, inference algorithms, and their applications to sequence modeling and causal reasoning.

    20 lessons
    1. Directed Graphical Models
    2. Undirected Graphical Models (MRFs)
    3. Variable Elimination
    4. Monte Carlo Sampling Methods
    5. Hidden Markov Models (HMMs)
    6. Conditional Independence and D-Separation
    7. Directed vs. Undirected Models
    8. Belief Propagation (Sum-Product Algorithm)
    9. Markov Chain Monte Carlo (MCMC)
    10. Forward-Backward and Viterbi Algorithms
    11. Parameter Learning in Bayesian Networks
    12. Factor Graphs
    13. The Junction Tree Algorithm
    14. Variational Inference
    15. Conditional Random Fields (CRFs)
    16. Structure Learning
    17. Energy-Based Models
    18. Computational Complexity of Inference
    19. Loopy Belief Propagation
    20. Causal Inference with Graphical Models
  7. Unit 7

    Unsupervised Learning

    Explore algorithms that discover hidden structure in unlabeled data, from clustering methods that group similar observations to dimensionality reduction techniques that reveal the essential structure of high-dimensional datasets. This course covers both classical techniques and modern visualization methods.

    20 lessons
    1. The K-Means Algorithm
    2. Agglomerative Hierarchical Clustering
    3. Gaussian Mixture Models (GMMs)
    4. Principal Component Analysis (PCA)
    5. Anomaly Detection Approaches
    6. Initialization Strategies: K-Means++
    7. Linkage Criteria and Their Effects
    8. The Expectation-Maximization (EM) Algorithm
    9. Explained Variance and Choosing Components
    10. Isolation Forest
    11. Choosing the Number of Clusters
    12. DBSCAN: Density-Based Clustering
    13. GMM Model Selection: BIC and AIC
    14. t-SNE: t-Distributed Stochastic Neighbor Embedding
    15. Association Rule Mining
    16. K-Medoids and K-Means Variants
    17. HDBSCAN: Hierarchical Density-Based Clustering
    18. Covariance Types and Their Impact
    19. UMAP: Uniform Manifold Approximation and Projection
    20. Evaluating Unsupervised Learning
  8. Unit 8

    Probability and Statistics for Machine Learning

    Develop a rigorous understanding of probability theory and statistical methods that underpin machine learning algorithms. From Bayes' theorem to information theory, this course equips you with the probabilistic reasoning skills essential for understanding generative models, Bayesian inference, and model evaluation.

    23 lessons
    1. Probability Axioms and Sample Spaces
    2. Random Variables and Their Types
    3. Maximum Likelihood Estimation (MLE)
    4. Null and Alternative Hypotheses
    5. Entropy
    6. Conditional Probability
    7. Expectation & Variance
    8. Maximum A Posteriori Estimation (MAP)
    9. P-Values and Statistical Significance
    10. Cross-Entropy
    11. Law of Total Probability
    12. Bernoulli & Binomial Distributions
    13. Bias and Variance of Estimators
    14. Confidence Intervals
    15. KL Divergence
    16. Independence and Conditional Independence
    17. The Gaussian (Normal) Distribution
    18. Consistency and Efficiency
    19. Type I and Type II Errors
    20. Mutual Information
    21. Bayes' Theorem
    22. Other Key Distributions: Poisson, Exponential, and Categorical
    23. Multiple Testing and Correction Methods
  9. Unit 9

    Data Thinking and Preparation

    Learn the foundational skills of working with data that determine the success or failure of any machine learning project. This course covers data types, collection strategies, cleaning techniques, feature engineering, and the critical practices that prevent data leakage and ensure reliable model evaluation.

    20 lessons
    1. Structured vs. Unstructured Data
    2. Data Collection Methods
    3. Missing Data Mechanisms
    4. Exploratory Data Analysis (EDA)
    5. Train, Validation, and Test Splits
    6. Numerical and Categorical Features
    7. Sampling Bias
    8. Imputation Strategies
    9. Normalization and Standardization
    10. Stratified and Temporal Splitting
    11. Ordinal Data
    12. Label Quality and Annotation
    13. Outlier Detection and Treatment
    14. Feature Engineering Basics
    15. Data Leakage
    16. Time Series Data
    17. Ethical Data Practices
    18. Deduplication and Data Consistency
    19. Encoding Categorical Features
    20. Leakage-Prevention Pipelines
  10. Unit 10

    Optimization and Search

    Explore the optimization algorithms and search strategies that enable machines to learn from data and solve complex problems. From gradient descent, the workhorse of modern ML, to classical search algorithms and metaheuristic approaches, this course covers the computational methods that find good solutions in vast search spaces.

    21 lessons
    1. Gradient Descent Fundamentals
    2. Convex Functions
    3. Constrained Optimization Formulations
    4. Breadth-First and Depth-First Search
    5. Simulated Annealing
    6. Batch, Mini-Batch, and Stochastic Gradient Descent
    7. Non-Convex Optimization in Deep Learning
    8. Lagrange Multipliers
    9. A* Search
    10. Genetic Algorithms
    11. The Learning Rate
    12. Saddle Points and Plateaus
    13. KKT Conditions
    14. Heuristic Design and Admissibility
    15. Particle Swarm Optimization
    16. Momentum and Nesterov Acceleration
    17. Convergence Conditions and Guarantees
    18. Primal-Dual Formulations
    19. Iterative Deepening and Bidirectional Search
    20. When to Use Metaheuristics
    21. Adaptive Learning Rate Methods: Adam, RMSProp, AdaGrad

Deep Learning & Modern Architectures

Neural networks from first principles through modern architectures — CNNs, transformers, generative models, reinforcement learning, and graph neural networks. Taught as mathematics and architecture rather than framework tutorials, so the knowledge survives tool churn.

  1. Unit 11

    Neural Networks from Scratch

    Build a deep understanding of neural networks by constructing them from first principles. Starting with the biological inspiration behind artificial neurons, you will derive the mathematics of the perceptron, compose multi-layer networks, implement forward propagation and backpropagation, and understand why neural networks are universal function approximators. This course bridges the gap between mathematical foundations and modern deep learning frameworks.

    25 lessons
    1. Biological Neurons and Artificial Inspiration
    2. The Role of Nonlinearity
    3. Forward Propagation Mechanics
    4. Chain Rule for Composite Functions
    5. Universal Approximation Theorem
    6. The Perceptron Model
    7. Sigmoid Activation Function
    8. Mean Squared Error Loss
    9. Backpropagation Derivation
    10. Depth vs Width Tradeoffs
    11. Perceptron Learning Rule
    12. Tanh Activation Function
    13. Cross-Entropy Loss
    14. Computational Graphs
    15. Why Initialization Matters
    16. Limitations and the XOR Problem
    17. ReLU and Its Variants
    18. Loss Surfaces and Optimization Landscape
    19. Gradient Flow and Debugging
    20. Xavier (Glorot) Initialization
    21. From Single Neurons to Multi-Layer Networks
    22. Softmax Activation Function
    23. Choosing and Combining Loss Functions
    24. Automatic Differentiation in Practice
    25. He (Kaiming) Initialization
  2. Unit 12

    Reinforcement Learning

    Learn how agents learn optimal behavior through interaction with environments. This course covers the mathematical framework of Markov decision processes, progresses through dynamic programming, Monte Carlo, and temporal difference methods, and culminates in modern deep reinforcement learning algorithms. You will understand the exploration-exploitation tradeoff, policy gradient methods, and the actor-critic architectures that have achieved superhuman performance in games and robotics.

    25 lessons
    1. Agent-Environment Interface
    2. Dynamic Programming for MDPs
    3. TD Prediction (TD(0))
    4. Policy Gradient Theorem
    5. Deep Q-Networks (DQN)
    6. Markov Decision Processes
    7. Value Iteration
    8. SARSA: On-Policy TD Control
    9. REINFORCE Algorithm
    10. Double DQN and Dueling DQN
    11. Value Functions
    12. Monte Carlo Prediction
    13. Q-Learning: Off-Policy TD Control
    14. Variance Reduction with Baselines
    15. Proximal Policy Optimization (PPO)
    16. Bellman Equations
    17. Exploration vs Exploitation
    18. TD(lambda) and Eligibility Traces
    19. Actor-Critic Architecture
    20. RL in Continuous Action Spaces
    21. Policies and Optimality
    22. Monte Carlo Control
    23. Reward Shaping
    24. Generalized Advantage Estimation (GAE)
    25. Challenges and Frontiers of Deep RL
  3. Unit 13

    Graph and Geometric Machine Learning

    Extend deep learning beyond grids and sequences to the rich world of graphs and geometric structures. This course covers how to apply neural networks to irregularly structured data like social networks, molecular graphs, and point clouds. Starting from graph theory fundamentals for ML, you will progress through message passing, graph neural network architectures, and the geometric deep learning framework that unifies CNNs, transformers, and GNNs under the principles of symmetry and equivariance.

    23 lessons
    1. Graph Fundamentals for Machine Learning
    2. Message Passing Framework
    3. Graph Convolutional Networks (GCN)
    4. Molecular Graphs and Drug Discovery
    5. Equivariance and Invariance
    6. Graph Representations for Neural Networks
    7. Aggregation Functions
    8. Graph Attention Networks (GAT)
    9. Social Network Analysis with GNNs
    10. Symmetry Groups in Machine Learning
    11. Graph Laplacian and Spectral Properties
    12. Multi-Layer Information Propagation
    13. Node, Edge, and Graph-Level Tasks
    14. Graph-Based Recommendation Systems
    15. Unifying CNNs, GNNs, and Transformers
    16. Challenges of Learning on Graphs
    17. Over-Smoothing Problem
    18. Graph Pooling and Readout Functions
    19. Point Clouds and 3D Data
    20. Equivariant Neural Networks
    21. Expressiveness and the WL Test
    22. Knowledge Graph Reasoning
    23. Frontiers of Geometric Machine Learning
  4. Unit 14

    Sequence Models and Attention

    Chart the evolution from recurrent neural networks to the transformer architecture that powers modern AI. Beginning with the fundamental challenge of modeling sequential data, this course progresses through RNNs, LSTMs, and GRUs, then pivots to the attention mechanism and the transformer architecture that replaced recurrence with parallelizable self-attention. You will understand the design decisions behind the most influential neural architecture of the current era.

    24 lessons
    1. RNN Fundamentals
    2. LSTM Architecture
    3. Attention Intuition and Motivation
    4. Positional Encoding
    5. Scaling Laws for Neural Language Models
    6. Backpropagation Through Time
    7. LSTM Gate Mechanics
    8. Query-Key-Value Framework
    9. Encoder-Decoder Transformer
    10. Encoder-Only Models (BERT Paradigm)
    11. Sequence-to-Sequence Framework
    12. GRU Architecture
    13. Scaled Dot-Product Attention
    14. Causal Masking
    15. Decoder-Only Models (GPT Paradigm)
    16. Teacher Forcing
    17. LSTM vs GRU: When to Use Which
    18. Self-Attention
    19. Feed-Forward Networks in Transformers
    20. Encoder-Decoder Models (T5 Paradigm)
    21. Bidirectional RNNs
    22. Multi-Head Attention
    23. Residual Connections and Layer Normalization in Transformers
    24. Efficient Attention Variants
  5. Unit 15

    Training Deep Networks

    Master the techniques that make training deep neural networks practical and reliable. From diagnosing and fixing gradient pathologies to selecting optimizers, normalization strategies, and regularization methods, this course covers the engineering craft behind successful deep learning. You will learn to tune learning rate schedules, apply mixed precision training for efficiency, and leverage transfer learning to bootstrap new models from pretrained ones.

    24 lessons
    1. Vanishing Gradients
    2. Internal Covariate Shift
    3. Overfitting in Deep Networks
    4. SGD with Momentum
    5. Mixed Precision Training Fundamentals
    6. Exploding Gradients
    7. Batch Normalization
    8. Dropout Regularization
    9. AdaGrad and RMSProp
    10. Loss Scaling for Mixed Precision
    11. Gradient Clipping
    12. Layer Normalization
    13. Early Stopping
    14. Adam Optimizer
    15. Transfer Learning Fundamentals
    16. Diagnosing Gradient and Activation Issues
    17. Group and Instance Normalization
    18. Weight Decay and L2 Regularization
    19. Learning Rate Warmup
    20. Fine-Tuning Strategies
    21. Normalization Placement Strategies
    22. Data Augmentation as Regularization
    23. Cosine Annealing and Other Schedules
    24. Feature Extraction vs Full Fine-Tuning
  6. Unit 16

    ML Engineering and Operations

    Bridge the gap between training models and running them reliably in production. This course covers the full machine learning lifecycle from experiment tracking and reproducibility through deployment, monitoring, and maintenance. You will learn the engineering practices, tools, and system design patterns that enable teams to develop, deploy, and iterate on ML systems with confidence and at speed.

    23 lessons
    1. The ML Lifecycle
    2. Data Versioning
    3. ML Testing Pyramid
    4. Model Monitoring Fundamentals
    5. ML System Design Patterns
    6. Experiment Tracking
    7. Model Registry and Versioning
    8. Continuous Training Pipelines
    9. Data Drift Detection
    10. Technical Debt in ML Systems
    11. Reproducibility in ML
    12. Feature Stores
    13. Model Validation Gates
    14. Concept Drift
    15. Data Management Best Practices
    16. Hyperparameter Management
    17. Data and Model Lineage
    18. ML Deployment Strategies
    19. A/B Testing for ML Models
    20. ML Governance and Documentation
    21. Rollback and Recovery
    22. Alerting and Incident Response
    23. ML Team Practices and Collaboration
  7. Unit 17

    Scalable ML Systems

    Learn how to train and deploy machine learning models at scale. This course covers the hardware, parallelism strategies, and optimization techniques that make it possible to train models with billions of parameters and serve them efficiently in production. From data parallelism and model sharding to quantization and knowledge distillation, you will understand the systems engineering that turns research models into practical, deployable AI systems.

    23 lessons
    1. GPU Architecture for Deep Learning
    2. Data Parallelism
    3. Anatomy of GPU Memory Usage
    4. Model Pruning
    5. Batch Serving vs Real-Time Serving
    6. TPU Architecture and Comparison
    7. Model Parallelism
    8. Activation Checkpointing (Gradient Checkpointing)
    9. Model Quantization
    10. Inference Optimization Techniques
    11. Memory and Bandwidth Bottlenecks
    12. Pipeline Parallelism
    13. ZeRO: Zero Redundancy Optimizer
    14. Knowledge Distillation
    15. Speculative Decoding
    16. Hardware Selection for ML Workloads
    17. Gradient Accumulation
    18. CPU and NVMe Offloading
    19. Compression Tradeoffs and Combining Techniques
    20. ML Serving Infrastructure
    21. 3D Parallelism and Hybrid Strategies
    22. Flash Attention and Memory-Efficient Attention
    23. Cost Optimization for ML Inference
  8. Unit 18

    Convolutional Neural Networks

    Explore the architecture that revolutionized computer vision. Starting from the convolution operation and its properties, this course builds up through pooling, feature maps, and the landmark architectures that pushed the boundaries of image recognition. You will understand how CNNs exploit spatial structure to learn hierarchical visual features, and master the design principles behind modern convolutional architectures.

    20 lessons
    1. Convolution Fundamentals
    2. Max Pooling
    3. LeNet: The Pioneer
    4. 1x1 Convolutions
    5. Filters and Kernels
    6. Average and Global Pooling
    7. AlexNet: The Deep Learning Revolution
    8. Depthwise Separable Convolutions
    9. Stride and Padding
    10. Feature Map Hierarchy
    11. VGGNet: Depth with Simplicity
    12. Dilated (Atrous) Convolutions
    13. Output Dimension Calculation
    14. Receptive Field
    15. ResNet: Skip Connections and Depth
    16. Transposed Convolutions
    17. Translation Equivariance and Weight Sharing
    18. Channels and Depth Dimension
    19. Architecture Evolution Principles
    20. CNN Architectural Design Principles
  9. Unit 19

    Generative Models

    Explore the architectures that learn to generate new data — images, text, audio, and beyond. This course covers the fundamental distinction between generative and discriminative modeling, then dives into autoencoders, variational autoencoders, generative adversarial networks, diffusion models, and autoregressive generation. You will understand the mathematical foundations, training dynamics, and evaluation metrics that define each approach.

    24 lessons
    1. Generative vs Discriminative Models
    2. Basic Autoencoders
    3. GAN Framework and Minimax Game
    4. Diffusion Model Intuition
    5. Frechet Inception Distance (FID)
    6. Modeling High-Dimensional Distributions
    7. Variational Autoencoders (VAE)
    8. Mode Collapse and Training Instability
    9. Denoising Score Matching
    10. Inception Score (IS)
    11. Latent Variable Models
    12. Evidence Lower Bound (ELBO)
    13. Wasserstein GAN (WGAN)
    14. Normalizing Flows
    15. Quality vs Diversity Tradeoff
    16. Maximum Likelihood for Generative Models
    17. The Reparameterization Trick
    18. Conditional GANs
    19. Classifier-Free Guidance
    20. Comparing Generative Model Families
    21. Latent Space Properties and Interpolation
    22. Progressive Training and StyleGAN
    23. Autoregressive Generative Models
    24. Applications of Generative Models
  10. Unit 20

    Representation Learning

    Understand how neural networks learn useful representations of data, from static word embeddings to the self-supervised paradigms that power modern foundation models. This course covers the progression from hand-crafted features to learned representations, the techniques for embedding discrete and continuous data into meaningful vector spaces, and the contrastive and predictive objectives that enable learning from unlabeled data at scale.

    23 lessons
    1. From Hand-Crafted Features to Learned Representations
    2. Contextual Embeddings
    3. Self-Supervised Learning Paradigm
    4. What is Disentanglement?
    5. Transfer Learning Theory and Practice
    6. Embedding Spaces
    7. Masked Language Modeling
    8. Contrastive Learning Framework
    9. Beta-VAE and Regularized Disentanglement
    10. Domain Adaptation
    11. Word2Vec
    12. Next-Token Prediction as Representation Learning
    13. SimCLR: Simple Contrastive Learning
    14. Measuring Disentanglement
    15. Foundation Models
    16. GloVe Embeddings
    17. Probing and Understanding Representations
    18. CLIP: Contrastive Language-Image Pretraining
    19. Theoretical Limits of Unsupervised Disentanglement
    20. Parameter-Efficient Fine-Tuning
    21. Limitations of Static Embeddings
    22. Non-Contrastive Self-Supervised Methods
    23. Risks and Limitations of Foundation Models

AI in the Real World

Applying AI to real domains (NLP, vision, recommender systems) and reasoning about its societal impact — safety, alignment, fairness, governance, human-AI interaction, and economics. These concerns only grow more important as AI capabilities increase.

  1. Unit 21

    AI Governance & Regulation

    Navigate the rapidly evolving landscape of AI governance, from landmark regulations like the EU AI Act to corporate responsible AI frameworks. This course examines how governments, international bodies, and industry are establishing rules for AI development and deployment, covering risk-based regulatory approaches, technical standards, intellectual property challenges, data privacy intersections, and the open source debate. You will develop the analytical skills to evaluate governance proposals and understand their implications for AI practitioners.

    21 lessons
    1. The EU AI Act
    2. Risk-Based AI Governance Frameworks
    3. Responsible AI Frameworks
    4. Copyright & AI Training Data
    5. Open Source vs Closed Models Debate
    6. US AI Policy & Executive Orders
    7. ISO & IEEE AI Standards
    8. AI Governance Organizational Structure
    9. Ownership of AI-Generated Content
    10. Foundation Model Governance
    11. China's AI Regulations
    12. AI Risk Assessment in Practice
    13. AI Lifecycle Governance
    14. GDPR & Machine Learning
    15. International AI Governance Cooperation
    16. Regulatory Approaches Compared
    17. Conformity Assessment & Certification
    18. AI Incident Response
    19. Privacy-Preserving ML Techniques
    20. Compute Governance
    21. AI & Patent Law
  2. Unit 22

    AI Safety & Alignment

    Confront the fundamental challenge of building AI systems that reliably do what we intend. This course covers the alignment problem from multiple angles: specification challenges like reward hacking and specification gaming, training-time alignment techniques including RLHF and constitutional AI, post-hoc interpretability methods, adversarial robustness, and the broader debate around existential risk. You will develop the conceptual toolkit to reason about AI safety as both a current engineering challenge and a long-term research frontier.

    23 lessons
    1. The Alignment Problem Defined
    2. Reinforcement Learning from Human Feedback (RLHF)
    3. LIME (Local Interpretable Model-Agnostic Explanations)
    4. Adversarial Examples
    5. Existential Risk Arguments
    6. Reward Hacking
    7. Constitutional AI
    8. SHAP (SHapley Additive exPlanations)
    9. Adversarial Attack Methods
    10. AI Containment & Boxing
    11. Specification Gaming
    12. AI Debate & Amplification
    13. Attention Visualization
    14. Adversarial Defense Strategies
    15. Corrigibility & Shutdown Problem
    16. Outer vs Inner Alignment
    17. Limitations of Reward Modeling
    18. Mechanistic Interpretability
    19. Distribution Shift & Out-of-Distribution Detection
    20. The Bridge from Current to Future Safety
    21. Mesa-Optimization
    22. Faithfulness vs Usefulness of Explanations
    23. Responsible AI Development Practices
  3. Unit 23

    AI Economics & Labor

    Analyze the profound economic transformations driven by artificial intelligence, from labor market disruption and productivity gains to the concentration of economic power and the global distribution of AI's benefits and costs. This course examines how AI reshapes work, creates and destroys jobs, alters competitive dynamics, and challenges existing economic institutions. You will develop the economic literacy to reason about AI's impact on employment, inequality, and prosperity.

    20 lessons
    1. AI in Developing Economies
    2. Historical Patterns of Technological Unemployment
    3. AI Productivity Gains
    4. Economics of Training Large Models
    5. AI & Economic Inequality
    6. AI Geopolitics & Competition
    7. Task-Based Framework for Automation
    8. Comparative Advantage with AI
    9. Data as an Economic Asset
    10. Universal Basic Income Debate
    11. The AI Digital Divide
    12. Evidence on AI Job Displacement
    13. The AI Productivity Paradox
    14. Winner-Take-All Dynamics
    15. Retraining & Education Policy
    16. AI Taxation Strategies
    17. The Future of Work with AI
    18. AI-Driven Job Creation
    19. Skill Premiums & Job Polarization
    20. AI Business Models
  4. Unit 24

    Computer Vision Applications

    Discover how convolutional and transformer-based architectures are applied to real-world visual understanding tasks. This course covers the major application areas of computer vision including image classification at scale, object detection, semantic and instance segmentation, generative image models, video understanding, 3D vision, and domain-specific applications in medical imaging and autonomous driving. You will learn the architectures, evaluation protocols, and deployment considerations that define production computer vision systems.

    23 lessons
    1. Modern Classification Architectures
    2. Two-Stage Detectors (R-CNN Family)
    3. Semantic Segmentation
    4. Image Generation with GANs
    5. Medical Imaging Applications
    6. Transfer Learning for Vision
    7. Single-Stage Detectors (YOLO & SSD)
    8. Instance Segmentation
    9. Diffusion Models for Image Synthesis
    10. Autonomous Driving Perception
    11. Data Augmentation Strategies
    12. Anchor-Free Detection
    13. Panoptic Segmentation
    14. Neural Style Transfer
    15. Depth Estimation
    16. Classification Evaluation & Error Analysis
    17. Detection Evaluation Metrics
    18. Segmentation Loss Functions
    19. Video Understanding & Temporal Modeling
    20. Point Cloud Processing
    21. Vision Transformers
    22. Feature Pyramid Networks
    23. Object Tracking
  5. Unit 25

    Fairness, Accountability & Transparency

    Develop a rigorous understanding of how AI systems can perpetuate or amplify societal biases and what technical and institutional mechanisms exist to address these harms. This course covers the mathematical definitions of fairness and their inherent tensions, methods for auditing algorithms, documentation practices like model cards and datasheets, and landmark case studies that illustrate how bias manifests in real-world AI deployments. You will learn to analyze, measure, and mitigate unfairness in machine learning systems.

    23 lessons
    1. Sources of Bias in ML Pipelines
    2. Demographic Parity
    3. Audit Methodology & Frameworks
    4. Model Cards
    5. COMPAS & Criminal Justice Bias
    6. Allocative vs Representational Harms
    7. Equalized Odds & Equal Opportunity
    8. Disparate Impact Analysis
    9. Datasheets for Datasets
    10. Hiring Algorithm Bias
    11. Bias Amplification
    12. Individual vs Group Fairness
    13. Subgroup Performance Evaluation
    14. Algorithmic Impact Assessments
    15. Pre-Processing Bias Mitigation
    16. Dataset Bias Case Studies
    17. Fairness Impossibility Theorems
    18. External & Independent Auditing
    19. Transparency & Explainability Requirements
    20. In-Processing & Post-Processing Mitigation
    21. Choosing Fairness Metrics in Practice
    22. Open Source & Reproducibility
    23. Organizational Fairness Strategy
  6. Unit 26

    Recommender Systems

    Learn how to build systems that predict what users want before they ask. This course covers the full landscape of recommendation algorithms from classical collaborative and content-based filtering through modern deep learning approaches, examining how platforms like Netflix, Amazon, and Spotify personalize content for billions of users. You will study matrix factorization, hybrid methods, evaluation metrics specific to ranking, exploration-exploitation trade-offs, and the ethical dimensions of algorithmic recommendation.

    22 lessons
    1. User-Based Collaborative Filtering
    2. Content-Based Filtering
    3. Ranking Metrics (NDCG & MAP)
    4. Neural Collaborative Filtering
    5. Filter Bubbles & Echo Chambers
    6. Item-Based Collaborative Filtering
    7. Matrix Factorization
    8. Precision@K and Recall@K
    9. Sequential & Session-Based Recommendations
    10. Popularity Bias
    11. Similarity Measures for Recommendations
    12. Hybrid Recommendation Approaches
    13. Offline Evaluation Protocols
    14. Graph Neural Networks for Recommendations
    15. Multi-Stakeholder Fairness
    16. Implicit vs Explicit Feedback
    17. Factorization Machines
    18. Beyond-Accuracy Metrics
    19. Multi-Armed Bandits for Exploration
    20. Responsible Recommendation Design
    21. The Cold Start Problem
    22. A/B Testing Recommender Systems
  7. Unit 27

    Human-AI Interaction

    Explore how humans and AI systems work together, focusing on the cognitive, social, and design factors that determine whether AI augments or undermines human capability. This course covers trust calibration, automation bias, human-AI teaming models, conversational AI design, accessibility considerations, and UX principles specific to AI-powered products. You will learn to design AI systems that are not just technically capable but genuinely usable and beneficial for the people who interact with them.

    21 lessons
    1. Cognitive Biases in AI Interaction
    2. Augmentation vs Automation
    3. Types of AI Explanations
    4. Conversational Design Principles
    5. AI-Created Accessibility Barriers
    6. Trust Calibration
    7. Task Allocation in Human-AI Teams
    8. Communicating Uncertainty
    9. AI Persona & Tone Design
    10. AI as Assistive Technology
    11. Mental Models of AI
    12. Collaborative Intelligence
    13. When & How to Explain
    14. Error Recovery & Graceful Degradation
    15. Inclusive AI Design Principles
    16. Automation Complacency & Skill Degradation
    17. Adaptive Autonomy
    18. Explanation Pitfalls & Dark Patterns
    19. Cognitive Load in AI-Assisted Work
    20. AI UX Design Principles
    21. Multimodal AI Interaction
  8. Unit 28

    Frontiers & Open Problems

    Survey the cutting edge of artificial intelligence research and the most important unsolved problems that define the field's future. This course covers the AGI debate, consciousness and AI, multimodal foundation models, neurosymbolic approaches, world models, AI for scientific discovery, quantum machine learning, and the environmental costs of AI. You will develop the skills to read and critically evaluate current ML research papers and understand where the field is heading.

    22 lessons
    1. AGI Definitions & Debates
    2. Multimodal Foundation Model Architectures
    3. Limitations of Pure Neural & Pure Symbolic AI
    4. AI for Scientific Discovery
    5. Energy & Environmental Costs of AI
    6. AGI Timeline Predictions
    7. World Models
    8. Neurosymbolic Integration Strategies
    9. AI for Mathematics & Theorem Proving
    10. Open Research Questions in AI
    11. Consciousness & AI
    12. Vision-Language Integration
    13. Program Synthesis & Reasoning
    14. Quantum Machine Learning
    15. How to Read ML Research Papers
    16. Understanding vs Computation
    17. Emergent Multimodal Capabilities
    18. Knowledge Graphs & LLM Integration
    19. AI for Weather & Climate Science
    20. Evaluating AI Claims & Benchmarks
    21. Abstract Reasoning & Abstraction Learning
    22. Pathways to Contributing to AI Research
  9. Unit 29

    Natural Language Understanding

    Explore how machines process, interpret, and generate human language. This course covers the full NLP pipeline from text preprocessing and tokenization through modern large language models, examining core tasks like sentiment analysis, named entity recognition, question answering, and summarization. You will develop a deep understanding of transformer-based architectures (BERT, GPT), prompt engineering principles, and the challenges of hallucination and grounding that define the current frontier of language AI.

    25 lessons
    1. Text Normalization
    2. Language Modeling Objectives
    3. Question Answering Systems
    4. BERT Architecture & Masked Language Modeling
    5. Prompt Engineering Principles
    6. Tokenization Strategies
    7. Sentiment Analysis
    8. Text Summarization
    9. GPT Architecture & Autoregressive Generation
    10. Chain-of-Thought Reasoning
    11. Subword Encoding (BPE & WordPiece)
    12. Named Entity Recognition
    13. Machine Translation
    14. Scaling Laws & Emergent Abilities
    15. Hallucination & Confabulation
    16. Word Embeddings
    17. Text Classification
    18. Decoding Strategies
    19. Instruction Tuning & RLHF
    20. Retrieval-Augmented Generation (RAG)
    21. Contextual Embeddings
    22. Sequence Labeling
    23. Generation Evaluation Metrics
    24. In-Context Learning & Few-Shot Prompting
    25. Grounding & Factual Verification
  10. Unit 30

    AI for Decision Making

    Examine how artificial intelligence supports and transforms decision-making across high-stakes domains. This course bridges decision theory and causal reasoning with practical AI applications in healthcare, finance, and criminal justice, while critically evaluating when AI decision support helps and when it causes harm. You will learn to apply utility functions, multi-criteria decision frameworks, counterfactual reasoning, and human-in-the-loop design principles that ensure AI augments rather than replaces sound human judgment.

    23 lessons
    1. Decision Making Under Uncertainty
    2. Correlation vs Causation in AI
    3. Clinical Decision Support Systems
    4. Algorithmic Trading
    5. Human-AI Decision Teaming
    6. Utility Functions & Preferences
    7. Counterfactual Reasoning
    8. AI-Assisted Diagnosis
    9. AI Credit Scoring
    10. Automation Bias
    11. Multi-Criteria Decision Making
    12. Treatment Effect Estimation
    13. AI for Treatment Recommendations
    14. AI Fraud Detection
    15. Levels of AI Autonomy
    16. Value of Information
    17. Causal Graphs & DAGs
    18. Ethics of Healthcare AI
    19. Criminal Justice Risk Assessment
    20. When NOT to Use AI
    21. Heterogeneous Treatment Effects
    22. Predictive Policing
    23. Decision Audit & Accountability

Explore more roadmaps

History
A working understanding of human history — from the first hominids to the global world we live in now. The trunk walks the throughline most people never got: how civilizations actually emerge, why empires rise and fall, and how the modern world was made. Clusters dive into the cities, classical mind, slave trade, revolutions, world wars, and how historians actually do their job.
Calculus
A working understanding of calculus — the mathematics of change and accumulation. The trunk gives you the throughline from limits and derivatives through the Fundamental Theorem to the borderlands where calculus hands off to analysis. Clusters dive into rigorous limits, integration techniques, multivariable and vector calculus, differential equations, series, and complex analysis.
Investing & Trading
How markets, brokers, and order books actually work — and what the evidence says about trying to beat them. The trunk covers market mechanics, asset classes, leverage, and the risk math every participant needs; clusters dive into microstructure, execution, the retail arenas, sizing math, psychology, manias, and strategy evidence.
Psychology
A working understanding of psychology — what the science actually claims about how minds, brains, and behavior fit together. The trunk gives you the orientation a literate professional needs (how psychology knows what it knows, the brain behind behavior, learning, memory, decision-making, emotion, personality, social influence, development, pathology, and what therapy does); clusters dive into research methods, perception, cognition mechanics, social influence tactics, the field guide of disorders, applied behavioral science, and the psychology of everyday life.
Cybersecurity
A working understanding of cybersecurity — how attackers operate, how defenders push back, and the moving parts of every modern security program. The trunk gives you the throughline from threat model to defense to response; clusters dive into cryptography, network defense, web application security, secure coding, penetration testing, incident response, cloud security, and governance.
Philosophy
A guided tour through philosophical thinking — how to reason carefully, evaluate arguments, navigate ethics, and engage with the great questions about knowledge, reality, mind, and society. The trunk gives you orientation across the whole field; clusters dive deep on logic, ethics, epistemology, metaphysics, mind, politics, and the history of ideas.

Frequently asked questions

How long is the AI & Machine Learning roadmap?

About 1000 hours in all: 670 lessons across 30 units.

What does the AI & Machine Learning roadmap cover?

30 units, starting with Mathematical Foundations for Machine Learning, What Is Artificial Intelligence, Model Evaluation and Selection and How Machines Learn, and 26 more.

How do I start the AI & Machine Learning roadmap?

Get the Mochivia app on the App Store and open this roadmap there.

Start AI & Machine Learning

Download on the App Store