The Variational Inference Book - Mastering Generative AI
The Variational Book
  • About
  • Topics
  • Author
  • Gallery

The Variational Inference Book

Master Generative AI with Derivations, Illustrations & Intuitive Explanations

📚 Coming Winter 2026 | Free Chapter Download


Get the Free Chapter Join the conversation on X

About the Book

The Variational Book Cover

One Comprehensive Guide to Generative AI

Develop world-class foundational machine learning expertise through a unified probabilistic modeling framework. This book covers everything from fundamental concepts to cutting-edge techniques in generative AI.

Whether you're an undergraduate student, graduate researcher, ML scientist, or technology professional, this book provides rigorous mathematical derivations paired with intuitive explanations and visual illustrations to build deep understanding.

  • Comprehensive coverage from basics to advanced topics
  • Mathematical rigor with intuitive explanations
  • Rich illustrations and visual aids
  • Practical implementations and examples
  • Perfect for students, researchers, and practitioners

Snapshot of Covered Topics

🔬 Foundations

  • Quantifying Uncertainty
  • Generative AI Introduction
  • Parameter Estimation
  • Exponential Families
  • Divergence Metrics
  • Latent Variable Models
  • Model Selection

📊 Monte Carlo Methods

  • Monte Carlo Approximation
  • Likelihood-ratio estimator
  • Reparameterization Trick
  • Control Variates
  • Black-box Variational Inference

📊 Surrogate Distributions

  • Auxiliary Design
  • Mixture of Gaussians
  • Expectation Maximization
  • Variational Mixture of Gaussians

📊 Variational Methods

  • Variational Inference
  • Mean-Field Approximations
  • Stochastic Variational Inference
  • Amortized Inference
  • Vector-Quantized Inference

🌊 Flow-Based Models

  • Normalizing Flows
  • Invertible Architectures
  • Continuous Normalizing Flows
  • Flow Matching
  • Rectifier Flows

âš¡ Energy-Based Models

  • Implicit Distributions
  • Generative Adversarial Networks (GANs)
  • GANs Training and Limitations
  • GANs Architecture
  • Conditional GANs
  • Wasserstein GANs

âš¡ Energy & Score-Based Models

  • Score Matching
  • Langevin Dynamics
  • Sliced Score Matching
  • Denoising Score Matching
  • Stein Gradient

🎨 Diffusion Models

  • DDPM & DDIM
  • Denoising Diffusion Models
  • Guided Diffusion
  • Latent Diffusion
  • Reverse Samplers in Diffusion
  • Distillation Diffusion
  • Discrete Diffusion

🎨 Transformers

  • Fundamentals of Transformers
  • Diffusion Transformers
  • Vector Quantized GAN
  • Visual Autoregressive Modeling

🧠 Applications

  • Data Modeling
  • Image Synthesis
  • Representation Learning
  • Text-to-Image Generation
  • Text-to-Video Generation

About the Author

Author Photo

Yuri Plotkin

Yuri Plotkin is an ML scientist specializing in generative AI, currently based in Los Angeles. With a background in Biomedical Engineering and years of research experience, Yuri brings a unique interdisciplinary perspective to machine learning.

His passion for making complex mathematical concepts accessible has driven him to create this comprehensive guide that bridges rigorous theory with intuitive understanding. Through years of research and practical application, Yuri has developed a teaching approach that helps readers develop deep, foundational expertise in generative AI.

Twitter/X Email

Generative AI Showcase

Explore examples of cutting-edge generative AI in action - the same techniques covered in detail throughout the book.

Prompt: A cat with a red sombrero learning to surf and reading a generative AI book!

Your browser does not support the video tag.

Prompt: A video of a cat learning generative AI!

📥 Get A Free Chapter

Subscribe to receive a free chapter plus updates on generative AI insights and techniques.



© 2025 Yuri Plotkin. All rights reserved.

Contact: author@thevariationalbook.com | @TheVariational