
Physics, Architectures, and Learning
Connect geometry, physics, causality, and uncertainty with the architectures and training methods used to learn world models at scale.
Part of World Models Handbook. The online chapters are free to read; the PDF edition is yours to keep, with future updates included.
Build the understanding to reason about world models.
Learn how a system can represent its environment and use predictions to make decisions. The handbook connects the underlying ideas to architectures, planning methods, and evaluation, helping you reason about what a world model can do and where its limits lie.
See what you’ll learnWhat you will learn
- Represent physical and causal structure without treating every regularity as a law.
- Compare recurrent, transformer, token, diffusion, and embedding-based models.
- Design data, objectives, and training procedures around the model's intended use.
Who this volume is for
- AI and robotics engineers, researchers, graduate students, reinforcement learning practitioners, and technically curious readers who want a coherent path from foundational state estimation and control to modern generative world models.
Prerequisites: The state, probability, and control foundations in Volume 1, plus familiarity with neural networks and gradient-based optimization.
Volume 2 PDF + code
Own this 637-page edition
Keep the PDF and companion notebooks for offline reading, with future updates included. The corresponding chapters stay free online.
- Carefully typeset PDF
- Companion notebooks included
- Future updates and errata included
- Online chapters remain free to read
Volume 2 of 4 · Available now
$29
one-timeVersion 2026.10.0 · 637 pages · secure checkout via Stripe
Delivered by email · Free updates included
Want the complete library?
Access to every volume, delivered as separate PDFs. Companion notebooks included. Save $37 vs. buying available volumes separately.
Get all-volume access · $79On Michael’s writing
Feedback on Michael’s writing
Genuine feedback, with names abbreviated for privacy.
These comments concern Michael’s online writing and other handbooks, not this new PDF edition.
Meet the community“I am enjoying reading your writings. I appreciate the effort you put into it. I want to drop a note of thanks. Thank you!”
“I would like to thank you for your work. Specifically, the chapter: 'LIME Explainability: Complete Guide to Local Interpretable Model-Agnostic Explanations' was extremely helpful for me in understanding LIME and communicating to colleagues. Your work is inspiring and valuable. Thank you for doing it.”
“I just came across your website where your books are available for free to read. Thank you so much for making this available to the public. This will help a lot of people.”
“I found your book Language AI Handbook's explanations of Transformer architectures and production deployment strategies to be exceptionally clear and insightful”
“Your writing feels like a treasure trove of great material. So, thank you for compiling everything on your website!”
“Really appreciate the way you have explained the concepts. Simple to understand, yet builds up knowledge exponentially, all while reinforcing it with examples repeatedly. I've read the last two chapters `74.Self-attention & 75.Q,K,V` and I've to admit, my concepts have never been clearer. You really are doing an amazing job breaking down maths and complex concepts like a story!”
“Thank you Michael. It's top!”
“I read a few of your blogs on NLP and oh WOW, they really are something, now I exactly know how all of the terms Entropy, Cross-entropy and how it is connected to Perplexity, will definitely read more of your blogs.”
“I came across your collection of online handbooks covering quantitative finance, AI, and machine learning. I wanted to reach out and thank you for making such high-quality material openly available. They are incredibly helpful for my CFA Level II review and for brushing up ahead of ML interviews.”
“Thanks so much for establishing such an amazing community. I appreciate the opportunity to be a learner and a contributor to the community.”
“Thanks for putting cool stuff out into the world.”
“I came across one of your books, Language AI Handbook, and man, I'm loving it. I've read many books but have never found such a comprehensive book on large models that covers everything from start to finish with such depth.”
Read all 12 reader reviews
“I would like to thank you for your work. Specifically, the chapter: 'LIME Explainability: Complete Guide to Local Interpretable Model-Agnostic Explanations' was extremely helpful for me in understanding LIME and communicating to colleagues. Your work is inspiring and valuable. Thank you for doing it.”
“Really appreciate the way you have explained the concepts. Simple to understand, yet builds up knowledge exponentially, all while reinforcing it with examples repeatedly. I've read the last two chapters `74.Self-attention & 75.Q,K,V` and I've to admit, my concepts have never been clearer. You really are doing an amazing job breaking down maths and complex concepts like a story!”
“Thanks so much for establishing such an amazing community. I appreciate the opportunity to be a learner and a contributor to the community.”
“I just came across your website where your books are available for free to read. Thank you so much for making this available to the public. This will help a lot of people.”
“I am enjoying reading your writings. I appreciate the effort you put into it. I want to drop a note of thanks. Thank you!”
“Thank you Michael. It's top!”
“Thanks for putting cool stuff out into the world.”
“I found your book Language AI Handbook's explanations of Transformer architectures and production deployment strategies to be exceptionally clear and insightful”
“Your writing feels like a treasure trove of great material. So, thank you for compiling everything on your website!”
“I came across your collection of online handbooks covering quantitative finance, AI, and machine learning. I wanted to reach out and thank you for making such high-quality material openly available. They are incredibly helpful for my CFA Level II review and for brushing up ahead of ML interviews.”
“I read a few of your blogs on NLP and oh WOW, they really are something, now I exactly know how all of the terms Entropy, Cross-entropy and how it is connected to Perplexity, will definitely read more of your blogs.”
“I came across one of your books, Language AI Handbook, and man, I'm loving it. I've read many books but have never found such a comprehensive book on large models that covers everything from start to finish with such depth.”
Exact contents
18 chapters across 3 parts
Every linked chapter is free to read online.
Part IV: Space, Physics, and Causality
- 01Geometry, Mapping, and 3D/4D State
Develops the topic through coordinate systems and camera geometry, depth, pose, and scene flow, SLAM, occupancy, and semantic maps, and neural fields and dynamic 3D representations.
- 02Physical Dynamics and Constraints
Develops the topic through rigid and deformable dynamics, contact, friction, and collision, conservation laws and symmetries, and physics priors and constraint enforcement.
- 03Causal Models and Interventions
Develops the topic through structural causal models, interventions and counterfactuals, causal representation learning, and confounding, invariance, and transportability.
- 04Structured Simulators and Neural Operators
Develops the topic through differentiable simulators, graph-based physical models, neural differential equations, and operator learning for fields and PDEs.
- 05Stochasticity and Uncertainty
Develops the topic through aleatoric and epistemic uncertainty, multimodal futures, ensembles, Bayesian models, and distributions, and calibration and uncertainty propagation.
- 06Compositionality and Systematic Generalization
Develops the topic through factorized mechanisms, recombining objects, relations, and skills, out-of-distribution composition, and measuring systematic generalization.
Part V: World-Model Architectures
- 01Recurrent State-Space Models
Develops the topic through deterministic and stochastic hidden state, recurrent state-space models, posterior inference and prior dynamics, and multi-step latent rollout.
- 02Transformers and Neural State-Space Architectures
Develops the topic through attention over trajectories, causal sequence modeling, long-context memory and efficient recurrence, and transformer and state-space hybrids.
- 03Autoregressive Token World Models
Develops the topic through visual, action, and multimodal tokenizers, next-token dynamics, spatiotemporal token layouts, and exposure bias and long-horizon generation.
- 04Diffusion and Flow World Models
Develops the topic through conditional denoising dynamics, diffusion over states and trajectories, flow matching and continuous-time generation, and controllability, latency, and sampling tradeoffs.
- 05Predictive Embeddings, JEPA, and Energy-Based Models
Develops the topic through predicting representations instead of pixels, joint-embedding predictive architectures, energy functions and compatibility learning, and collapse prevention and target encoders.
- 06Hierarchical, Hybrid, and Omnimodal Models
Develops the topic through multiple temporal and spatial scales, neural-symbolic and simulator hybrids, shared state across modalities, and modular experts and routing.
Part VI: Learning World Models at Scale
- 01Predictive and Generative Objectives
Develops the topic through reconstruction and likelihood objectives, contrastive and masked prediction, multi-step and overshooting losses, and balancing representation, dynamics, and reward losses.
- 02Actions, Inverse Dynamics, and Latent Actions
Develops the topic through action-conditioned transitions, inverse dynamics and controllability, learning actions from unlabeled video, and latent action alignment with real controls.
- 03Interaction Data and Passive Observation
Develops the topic through agent-collected trajectories, observational video and internet data, coverage, behavior policy, and interventions, and combining passive and embodied experience.
- 04Data Curation and Synthetic Data
Develops the topic through trajectory quality and diversity, deduplication, filtering, and balancing, simulation and procedural generation, and data mixtures, provenance, and contamination.
- 05Online, Offline, and Continual Learning
Develops the topic through batch and streaming updates, distribution shift and replay, catastrophic forgetting, and continual adaptation without unsafe drift.
- 06Pretraining, Post-Training, Scaling, and Adaptation
Develops the topic through scaling data, parameters, context, and compute, task adaptation and instruction conditioning, preference and control-oriented post-training, and transfer across embodiments and domains.
Choose the edition that fits your next step
Volume 2: Physics, Architectures, and Learning
637 pages focused on this subject, with future updates included.
$29
Choose this volumeThe complete World Models Handbook library
All-volume access with separate volume PDFs and future updates. No combined PDF.
$79
Get all-volume accessExplore the other volumes

Volume 1 · 608 pages
Foundations, Control, and Representations
The World-Model Idea · Inference, Decisions, and Control · Representing Agents and Worlds

Volume 3 · 701 pages
Planning, Research Lineages, and Foundation Models
Planning and Agency · Decision-Centric Research Lineages · Foundation and World-Action Models

Volume 4 · 713 pages
Applications, Evaluation, and Reliable Deployment
Application Domains · Evaluation and Understanding · Reliable World Models
Author and edition details
About the author and this edition

Michael Brenndoerfer
Michael has spent more than a decade working across software engineering, data, AI, and business. He writes to understand difficult ideas more deeply and to share what he learns in a clear, practical way.
- Edition
- First edition
- Published
- Last reviewed
Version history
Kept current, not frozen in time
Library access includes every future volume and update. When a volume changes, its updated PDF appears in My books at no extra cost.
Current release
Edition 2026.10.0
First edition: 72 chapters across four standalone volume PDFs, with executable companion notebooks and perpetual updates.