The Variational Inference Book

The topic of generative AI unites key concepts in machine learning and provides a common framework for probabilistic modeling and inference. Most state-of-the art artificial intelligence algorithms (latent models, VAE, Gans, Normalizing Flows, Diffusion) leverage ideas from the topic.

Develop world-class foundational machine learning expertise

Everything in one concise, explanatory book.

Introduction to Generative AI
Uncertainty and Posterior Distributions
Bayesian inference and Maximum Likelihood
Latent Variable Models
Model selection and comparison
Exponential family distributions
KL-divergence
Evidence of Lower Bound (ELBO)
Mean-Field approximations

Snapshot of covered topics:

Stochastic Variational Inference
Amortized inference (VAE)
Monte-Carlo approximations
Variance reduction
Structured approximations
Normalizing Flows
Flow Matching
Auxillary distributions
Implicit (Energy) approximation

Generative Adverserial Networks (GANS)
Score Matching
Diffusion Models
Stochastic and Ordinary Equations
Reinforcement Learning

Who should read this book?

Undergraduate or graduate students studying computer science, data science, machine learning or STEM requiring a strong mathematical foundation in generative AI.
ML scientists and engineers, data scientists or software engineers interested in understanding machine learning and generative AI concepts and intuition without adding needless complexity (for grasping the ideas quickly).

Technology industry professionals poised to collaborate closely with generative AI tools.
Self-learner, go-getters and curious minds craving to master what machine learning is all about, including learning how generative AI tools are designed for creative needs, i.e. text2image, text2video models

Flux text2image generation

OpenAI Sora text2video generation

Many topics with in-depth explanations

This is a custom code placeholder.
Switch to Preview or publish the page
to see how your code works. Double-click to edit

<getresponse-form form-id="1146c153-635d-4dfc-a4a9-4e11537b5519" e="1"></getresponse-form>

About the author

Yuri Plotkin

Hi there! My name is Yuri. I'm a machine learning scientist currently living in Los Angeles. I received both my degrees in Biomedical Engineering and have spent time in the wet-lab. At the moment, my professional endeavors center around machine learning, primarily on generative AI. In my spare time, I enjoy reading computer science with a keen eye on variational, Bayesian, diffusion, and llm topics. I hope you enjoy reading the book as much as I have enjoyed writing it.

Feel free to reach me at Twitter and Email

FAQ

When will the book be available?

How does this book set itself apart from existing ML books?

The book release is planned for winter 2024.

Everything you need to know in one concise book, it covers all generative AI techniques used in modern machine learning algorithms. Read the book to grasp the mathematical principles and intuition required for being adept at all things AI.

What if I have a question?

Send an email to author@thevariationalbook.com