Part I: Artificial Intelligence

1. 1Introduction
2. 2Intelligent Agents

Part II: Problem-solving

1. 3Solving Problems by Searching
2. 4Search in Complex Environments
3. 5Adversarial Search and Games
4. 6Constraint Satisfaction Problems

Part III: Knowledge, Reasoning, and Planning

1. 7Logical Agents
2. 8First-Order Logic
3. 9Inference in First-Order Logic
4. 10Knowledge Representation
5. 11Automated Planning

Part IV: Uncertain Knowledge and Reasoning

1. 12Quantifying Uncertainty
2. 13Probabilistic Reasoning
3. 14Probabilistic Reasoning over Time
4. 15Probabilistic Programming
5. 16Making Simple Decisions
6. 17Making Complex Decisions
7. 18Multiagent Decision Making

Part V: Machine Learning

1. 19Learning from Examples
2. 20Learning Probabilistic Models
3. 21Deep Learning
4. 22Reinforcement Learning

Part VI: Communicating, Perceiving, and Acting

1. 23Natural Language Processing
2. 24Deep Learning for Natural Language Processing
3. 25Computer Vision
4. 26Robotics

Part VII: Conclusions

1. 27Philosophy, Ethics, and Safety of AI
2. 28The Future of AI

Part I: Agents in the World

1. 1Artificial Intelligence and Agents
2. 2Agent Architectures and Hierarchical Control

Part II: Reasoning and Planning with Certainty

1. 3Searching for Solutions
2. 4Reasoning with Constraints
3. 5Propositions and Inference
4. 6Deterministic Planning

Part III: Learning and Reasoning with Uncertainty

1. 7Supervised Machine Learning
2. 8Neural Networks and Deep Learning
3. 9Reasoning with Uncertainty
4. 10Learning with Uncertainty
5. 11Causality

Part IV: Planning and Acting with Uncertainty

1. 12Planning with Uncertainty
2. 13Reinforcement Learning
3. 14Multiagent Systems

Part V: Representing Individuals and Relations

1. 15Individuals and Relations
2. 16Knowledge Graphs and Ontologies
3. 17Relational Learning and Probabilistic Reasoning

Part VI: The Big Picture

1. 18The Social Impact of Artificial Intelligence
2. 19Retrospect and Prospect

1. ·Prologue: Terrified

Part I: Background

1. 1The Roots of Artificial Intelligence
2. 2Neural Networks and the Ascent of Machine Learning
3. 3AI Spring

Part II: Looking and Learning

1. 4Who, What, When, Where, Why
2. 5ConvNets and ImageNet
3. 6A Closer Look at Machines That Learn
4. 7On Trustworthy and Ethical AI

Part III: Learning to Play

1. 8Rewards for Robots
2. 9Game On
3. 10Beyond Games

Part IV: Artificial Intelligence Meets Natural Language

1. 11Words, and the Company They Keep
2. 12Translation as Encoding and Decoding
3. 13Ask Me Anything

Part V: The Barrier of Meaning

1. 14On Understanding
2. 15Knowledge, Abstraction, and Analogy in Artificial Intelligence
3. ·Epilogue: Questions, Answers, and Speculations

1. 1Introduction
2. 2Probability Distributions
3. 3Linear Models for Regression
4. 4Linear Models for Classification
5. 5Neural Networks
6. 6Kernel Methods
7. 7Sparse Kernel Machines
8. 8Graphical Models
9. 9Mixture Models and EM
10. 10Approximate Inference
11. 11Sampling Methods
12. 12Continuous Latent Variables
13. 13Sequential Data
14. 14Combining Models

1. 1Introduction

Part I: Foundations

1. 2Probability: Univariate Models
2. 3Probability: Multivariate Models
3. 4Statistics
4. 5Decision Theory
5. 6Information Theory
6. 7Linear Algebra
7. 8Optimization

Part II: Linear Models

1. 9Linear Discriminant Analysis
2. 10Logistic Regression
3. 11Linear Regression
4. 12Generalized Linear Models

Part III: Deep Neural Networks

1. 13Neural Networks for Tabular Data
2. 14Neural Networks for Images
3. 15Neural Networks for Sequences

Part IV: Nonparametric Models

1. 16Exemplar-based Methods
2. 17Kernel Methods
3. 18Trees, Forests, Bagging, and Boosting

Part V: Beyond Supervised Learning

1. 19Learning with Fewer Labeled Examples
2. 20Dimensionality Reduction
3. 21Clustering
4. 22Recommender Systems
5. 23Graph Embeddings

1. 1Introduction
2. 2Overview of Supervised Learning
3. 3Linear Methods for Regression
4. 4Linear Methods for Classification
5. 5Basis Expansions and Regularization
6. 6Kernel Smoothing Methods
7. 7Model Assessment and Selection
8. 8Model Inference and Averaging
9. 9Additive Models, Trees, and Related Methods
10. 10Boosting and Additive Trees
11. 11Neural Networks
12. 12Support Vector Machines and Flexible Discriminants
13. 13Prototype Methods and Nearest-Neighbors
14. 14Unsupervised Learning
15. 15Random Forests
16. 16Ensemble Learning
17. 17Undirected Graphical Models
18. 18High-Dimensional Problems: p >> N

1. 1Introduction
2. 2Statistical Learning
3. 3Linear Regression
4. 4Classification
5. 5Resampling Methods
6. 6Linear Model Selection and Regularization
7. 7Moving Beyond Linearity
8. 8Tree-Based Methods
9. 9Support Vector Machines
10. 10Deep Learning
11. 11Survival Analysis and Censored Data
12. 12Unsupervised Learning
13. 13Multiple Testing

Part I: The Fundamentals of Machine Learning

1. 1The Machine Learning Landscape
2. 2End-to-End Machine Learning Project
3. 3Classification
4. 4Training Models
5. 5Decision Trees
6. 6Ensemble Learning and Random Forests
7. 7Dimensionality Reduction
8. 8Unsupervised Learning Techniques

Part II: Neural Networks and Deep Learning

1. 9Introduction to Artificial Neural Networks
2. 10Building Neural Networks with PyTorch
3. 11Training Deep Neural Networks
4. 12Deep Computer Vision Using Convolutional Neural Networks
5. 13Processing Sequences Using RNNs and CNNs
6. 14Natural Language Processing with RNNs and Attention
7. 15Transformers for Natural Language Processing and Chatbots
8. 16Vision and Multimodal Transformers
9. 17Speeding Up Transformers
10. 18Autoencoders, GANs, and Diffusion Models
11. 19Reinforcement Learning
12. ·Appendix A. Autodiff
13. ·Appendix B. Mixed Precision and Quantization

1. 1Introduction
2. 2Concept Learning and the General-to-Specific Ordering
3. 3Decision Tree Learning
4. 4Artificial Neural Networks
5. 5Evaluating Hypotheses
6. 6Bayesian Learning
7. 7Computational Learning Theory
8. 8Instance-Based Learning
9. 9Genetic Algorithms
10. 10Learning Sets of Rules
11. 11Analytical Learning
12. 12Combining Inductive and Analytical Learning
13. 13Reinforcement Learning

1. 1Introduction

Part I: Foundations

1. 2A Gentle Start
2. 3A Formal Learning Model
3. 4Learning via Uniform Convergence
4. 5The Bias-Complexity Tradeoff
5. 6The VC-Dimension
6. 7Nonuniform Learnability
7. 8The Runtime of Learning

Part II: From Theory to Algorithms

1. 9Linear Predictors
2. 10Boosting
3. 11Model Selection and Validation
4. 12Convex Learning Problems
5. 13Regularization and Stability
6. 14Stochastic Gradient Descent
7. 15Support Vector Machines
8. 16Kernel Methods
9. 17Multiclass, Ranking, and Complex Prediction Problems
10. 18Decision Trees
11. 19Nearest Neighbor
12. 20Neural Networks

Part III: Additional Learning Models

1. 21Online Learning
2. 22Clustering
3. 23Dimensionality Reduction
4. 24Generative Models
5. 25Feature Selection and Generation

Part IV: Advanced Theory

1. 26Rademacher Complexities
2. 27Covering Numbers
3. 28Proof of the Fundamental Theorem of Learning Theory
4. 29Multiclass Learnability
5. 30Compression Bounds
6. 31PAC-Bayes

Part I: Mathematical Foundations

1. 1Introduction and Motivation
2. 2Linear Algebra
3. 3Analytic Geometry
4. 4Matrix Decompositions
5. 5Vector Calculus
6. 6Probability and Distributions
7. 7Continuous Optimization

Part II: Central Machine Learning Problems

1. 8When Models Meet Data
2. 9Linear Regression
3. 10Dimensionality Reduction with Principal Component Analysis
4. 11Density Estimation with Gaussian Mixture Models
5. 12Classification with Support Vector Machines

1. 1Introduction

Part I: Applied Math and Machine Learning Basics

1. 2Linear Algebra
2. 3Probability and Information Theory
3. 4Numerical Computation
4. 5Machine Learning Basics

Part II: Deep Networks: Modern Practices

1. 6Deep Feedforward Networks
2. 7Regularization for Deep Learning
3. 8Optimization for Training Deep Models
4. 9Convolutional Networks
5. 10Sequence Modeling: Recurrent and Recursive Nets
6. 11Practical Methodology
7. 12Applications

Part III: Deep Learning Research

1. 13Linear Factor Models
2. 14Autoencoders
3. 15Representation Learning
4. 16Structured Probabilistic Models for Deep Learning
5. 17Monte Carlo Methods
6. 18Confronting the Partition Function
7. 19Approximate Inference
8. 20Deep Generative Models

1. 1The Deep Learning Revolution
2. 2Probabilities
3. 3Standard Distributions
4. 4Single-layer Networks: Regression
5. 5Single-layer Networks: Classification
6. 6Deep Neural Networks
7. 7Gradient Descent
8. 8Backpropagation
9. 9Regularization
10. 10Convolutional Networks
11. 11Structured Distributions
12. 12Transformers
13. 13Graph Neural Networks
14. 14Sampling
15. 15Discrete Latent Variables
16. 16Continuous Latent Variables
17. 17Generative Adversarial Networks
18. 18Normalizing Flows
19. 19Autoencoders
20. 20Diffusion Models

1. 1Introduction
2. 2Supervised Learning
3. 3Shallow Neural Networks
4. 4Deep Neural Networks
5. 5Loss Functions
6. 6Fitting Models
7. 7Gradients and Initialization
8. 8Measuring Performance
9. 9Regularization
10. 10Convolutional Networks
11. 11Residual Networks
12. 12Transformers
13. 13Graph Neural Networks
14. 14Unsupervised Learning
15. 15Generative Adversarial Networks
16. 16Normalizing Flows
17. 17Variational Autoencoders
18. 18Diffusion Models
19. 19Reinforcement Learning
20. 20Why Does Deep Learning Work?
21. 21Deep Learning and Ethics

1. ·Installation and Notation
2. 1Introduction
3. 2Preliminaries
4. 3Linear Neural Networks for Regression
5. 4Linear Neural Networks for Classification
6. 5Multilayer Perceptrons
7. 6Builders' Guide
8. 7Convolutional Neural Networks
9. 8Modern Convolutional Neural Networks
10. 9Recurrent Neural Networks
11. 10Modern Recurrent Neural Networks
12. 11Attention Mechanisms and Transformers
13. ·Appendix: Tools for Deep Learning

1. 1What Is Deep Learning?
2. 2The Mathematical Building Blocks of Neural Networks
3. 3Introduction to TensorFlow, PyTorch, JAX, and Keras
4. 4Classification and Regression
5. 5Fundamentals of Machine Learning
6. 6The Universal Workflow of Machine Learning
7. 7A Deep Dive on Keras
8. 8Image Classification
9. 9ConvNet Architecture Patterns
10. 10Interpreting What ConvNets Learn
11. 11Image Segmentation
12. 12Object Detection
13. 13Timeseries Forecasting
14. 14Text Classification
15. 15Language Models and the Transformer
16. 16Text Generation
17. 17Image Generation
18. 18Best Practices for the Real World
19. 19The Future of AI
20. 20Conclusions

1. 1Introduction

Part I: Tabular Solution Methods

1. 2Multi-armed Bandits
2. 3Finite Markov Decision Processes
3. 4Dynamic Programming
4. 5Monte Carlo Methods
5. 6Temporal-Difference Learning
6. 7n-step Bootstrapping
7. 8Planning and Learning with Tabular Methods

Part II: Approximate Solution Methods

1. 9On-policy Prediction with Approximation
2. 10On-policy Control with Approximation
3. 11Off-policy Methods with Approximation
4. 12Eligibility Traces
5. 13Policy Gradient Methods

Part III: Looking Deeper

1. 14Psychology
2. 15Neuroscience
3. 16Applications and Case Studies
4. 17Frontiers

1. 1Introduction
2. 2Foundations

Part I: Representation

1. 3The Bayesian Network Representation
2. 4Undirected Graphical Models
3. 5Local Probabilistic Models
4. 6Template-Based Representations
5. 7Gaussian Network Models
6. 8The Exponential Family

Part II: Inference

1. 9Exact Inference: Variable Elimination
2. 10Exact Inference: Clique Trees
3. 11Inference as Optimization
4. 12Particle-Based Approximate Inference
5. 13MAP Inference
6. 14Inference in Hybrid Networks
7. 15Inference in Temporal Models

Part III: Learning

1. 16Learning Graphical Models: Overview
2. 17Parameter Estimation
3. 18Structure Learning in Bayesian Networks
4. 19Partially Observed Data
5. 20Learning Undirected Models

Part IV: Actions and Decisions

1. 21Causality
2. 22Utilities and Decisions
3. 23Structured Decision Problems
4. 24Epilogue

1. 1Introduction to Probabilities, Graphs, and Causal Models
2. 2A Theory of Inferred Causation
3. 3Causal Diagrams and the Identification of Causal Effects
4. 4Actions, Plans, and Direct Effects
5. 5Causality and Structural Models in Social Science and Economics
6. 6Simpson's Paradox, Confounding, and Collapsibility
7. 7The Logic of Structure-Based Counterfactuals
8. 8Imperfect Experiments: Bounding Effects and Counterfactuals
9. 9Probability of Causation: Interpretation and Identification
10. 10The Actual Cause
11. 11Reflections, Elaborations, and Discussions with Readers

1. ·Introduction: Mind over Data
2. 1The Ladder of Causation
3. 2From Buccaneers to Guinea Pigs: The Genesis of Causal Inference
4. 3From Evidence to Causes: Reverend Bayes Meets Mr. Holmes
5. 4Confounding and Deconfounding: Or, Slaying the Lurking Variable
6. 5The Smoke-Filled Debate: Clearing the Air
7. 6Paradoxes Galore!
8. 7Beyond Adjustment: The Conquest of Mount Intervention
9. 8Counterfactuals: Mining Worlds That Could Have Been
10. 9Mediation: The Search for a Mechanism
11. 10Big Data, Artificial Intelligence, and the Big Questions

1. 1Hello Transformers
2. 2Text Classification
3. 3Transformer Anatomy
4. 4Multilingual Named Entity Recognition
5. 5Text Generation
6. 6Summarization
7. 7Question Answering
8. 8Making Transformers Efficient in Production
9. 9Dealing with Few to No Labels
10. 10Training Transformers from Scratch
11. 11Future Directions

1. 1Understanding Large Language Models
2. 2Working with Text Data
3. 3Coding Attention Mechanisms
4. 4Implementing a GPT Model from Scratch to Generate Text
5. 5Pretraining on Unlabeled Data
6. 6Fine-tuning for Classification
7. 7Fine-tuning to Follow Instructions
8. ·Appendix A. Introduction to PyTorch
9. ·Appendix B. References and Further Reading
10. ·Appendix C. Exercise Solutions
11. ·Appendix D. Adding Bells and Whistles to the Training Loop
12. ·Appendix E. Parameter-efficient Fine-tuning with LoRA

Part I: Understanding Language Models

1. 1An Introduction to Large Language Models
2. 2Tokens and Embeddings
3. 3Looking Inside Large Language Models

Part II: Using Pretrained Language Models

1. 4Text Classification
2. 5Text Clustering and Topic Modeling
3. 6Prompt Engineering
4. 7Advanced Text Generation Techniques and Tools
5. 8Semantic Search and Retrieval-Augmented Generation
6. 9Multimodal Large Language Models

Part III: Training and Fine-Tuning Language Models

1. 10Creating Text Embedding Models
2. 11Fine-Tuning Representation Models for Classification
3. 12Fine-Tuning Generation Models

1. 1Introduction to Building AI Applications with Foundation Models
2. 2Understanding Foundation Models
3. 3Evaluation Methodology
4. 4Evaluate AI Systems
5. 5Prompt Engineering
6. 6RAG and Agents
7. 7Finetuning
8. 8Dataset Engineering
9. 9Inference Optimization
10. 10AI Engineering Architecture and User Feedback

Part I: Introduction to Generative Deep Learning

1. 1Generative Modeling
2. 2Deep Learning

Part II: Methods

1. 3Variational Autoencoders
2. 4Generative Adversarial Networks
3. 5Autoregressive Models
4. 6Normalizing Flow Models
5. 7Energy-Based Models
6. 8Diffusion Models

Part III: Applications

1. 9Transformers
2. 10Advanced GANs
3. 11Music Generation
4. 12World Models
5. 13Multimodal Models
6. 14Conclusion

1. 1If We Succeed
2. 2Intelligence in Humans and Machines
3. 3How Might AI Progress in the Future?
4. 4Misuses of AI
5. 5Overly Intelligent AI
6. 6The Not-So-Great AI Debate
7. 7AI: A Different Approach
8. 8Provably Beneficial AI
9. 9Complications: Us
10. 10Problem Solved?
11. ·Appendices A-D: Searching for Solutions; Knowledge and Logic; Uncertainty and Probability; Learning from Experience

1. ·Prologue
2. ·Introduction

I. Prophecy

1. 1Representation
2. 2Fairness
3. 3Transparency

II. Agency

1. 4Reinforcement
2. 5Shaping
3. 6Curiosity

III. Normativity

1. 7Imitation
2. 8Inference
3. 9Uncertainty
4. ·Conclusion

1. 1Past Developments and Present Capabilities
2. 2Paths to Superintelligence
3. 3Forms of Superintelligence
4. 4The Kinetics of an Intelligence Explosion
5. 5Decisive Strategic Advantage
6. 6Cognitive Superpowers
7. 7The Superintelligent Will
8. 8Is the Default Outcome Doom?
9. 9The Control Problem
10. 10Oracles, Genies, Sovereigns, Tools
11. 11Multipolar Scenarios
12. 12Acquiring Values
13. 13Choosing the Criteria for Choosing
14. 14The Strategic Picture
15. 15Crunch Time

1. ·Introduction
2. 1Earth
3. 2Labor
4. 3Data
5. 4Classification
6. 5Affect
7. 6State
8. ·Conclusion: Power
9. ·Coda: Space

1. ·Introduction
2. 1Bomb Parts: What Is a Model?
3. 2Shell Shocked: My Journey of Disillusionment
4. 3Arms Race: Going to College
5. 4Propaganda Machine: Online Advertising
6. 5Civilian Casualties: Justice in the Age of Big Data
7. 6Ineligible to Serve: Getting a Job
8. 7Sweating Bullets: On the Job
9. 8Collateral Damage: Landing Credit
10. 9No Safe Zone: Getting Insurance
11. 10The Targeted Citizen: Civic Life
12. ·Conclusion

1. 1Introduction
2. 2How Predictive AI Goes Wrong
3. 3Why Can't AI Predict the Future?
4. 4The Long Road to Generative AI
5. 5Is Advanced AI an Existential Threat?
6. 6Why Can't AI Fix Social Media?
7. 7Why Do Myths about AI Persist?
8. 8Where Do We Go from Here?

1. ·Introduction: Three Sleepless Nights

Part I

1. 1Creating Alien Minds
2. 2Aligning the Alien
3. 3Four Rules for Co-Intelligence

Part II

1. 4AI as a Person
2. 5AI as a Creative
3. 6AI as a Coworker
4. 7AI as a Tutor
5. 8AI as a Coach
6. 9AI as Our Future
7. ·Epilogue: AI as Us

1. 1Introduction: Machine Intelligence
2. 2Cheap Changes Everything

Part One: Prediction

1. 3Prediction Machine Magic
2. 4Why It's Called Intelligence
3. 5Data Is the New Oil
4. 6The New Division of Labor

Part Two: Decision Making

1. 7Unpacking Decisions
2. 8The Value of Judgment
3. 9Predicting Judgment
4. 10Taming Complexity
5. 11Fully Automated Decision Making

Part Three: Tools

1. 12Deconstructing Work Flows
2. 13Decomposing Decisions
3. 14Job Redesign

Part Four: Strategy

1. 15AI in the C-Suite
2. 16When AI Transforms Your Business
3. 17Your Learning Strategy
4. 18Managing AI Risk

Part Five: Society

1. 19Beyond Business

1. ·Preface: Colliding Worlds
2. 1The Age of AI
3. 2Rethinking the Firm
4. 3The AI Factory
5. 4Rearchitecting the Firm
6. 5Becoming an AI Company
7. 6Strategy for a New Age
8. 7Strategic Collisions
9. 8The Ethics of Digital Scale
10. 9The New Meta
11. 10A Leadership Mandate