Table of contents

Modern Neural Networks

From first principles to large language models and frontier research

Read each section in order. Every title can be opened as a TheoryTrace document.

• Cover
• Copyright
• How to read this book
• Introduction
• Chapter 1: The Computational Neuron
• Chapter 2: Multilayer Networks and Representation Learning
• Chapter 3: Optimization by Gradient Descent
• Chapter 4: Generalization, Regularization, and Model Capacity
• Chapter 5: Initialization, Normalization, and Training Dynamics
• Chapter 6: Convolutional Neural Networks
• Chapter 7: Sequence Models Before Transformers
• Chapter 8: Attention and the Transformer
• Chapter 9: Probabilistic Modeling and Neural Generative Models
• Chapter 10: Self-Supervised Learning and Foundation Models
• Chapter 11: Large Language Models
• Chapter 12: Scaling Laws and Emergent Capabilities
• Chapter 13: Instruction Tuning, Alignment, and Preference Optimization
• Chapter 14: Retrieval, Tools, and Agentic Systems
• Chapter 15: Efficient Training and Inference
• Chapter 16: Multimodal Neural Networks
• Chapter 17: Robustness, Interpretability, and Mechanistic Understanding
• Chapter 18: Evaluation, Benchmarks, and Scientific Validity
• Chapter 19: Safety, Ethics, and Societal Impact
• Chapter 20: Frontier Research and Open Problems
• Conclusion