Oreilly – Causal AI, Video Edition 2025-2

Updated August 10, 2026 2.1 GB

Download

About this item

Description

Causal AI, Video Edition. This course teaches how to build AI models that can reliably make causal inferences. Causal AI provides the insight needed to predict and control outcomes based on causal relationships rather than just correlations, enabling precise and timely interventions. As a practical introduction, this course covers building models that can reason about causality. While traditional machine learning models are unable to answer causal questions such as “Why did that happen?” or “What factors need to change to change the outcome?”, this course combines advanced statistical methods, computational techniques, and novel algorithms to create systems that automate the process of causal inference. This course introduces tools, techniques, and algorithms for causal reasoning for machine learning, skillfully combining Bayesian and probabilistic approaches with practical examples in Python. Author Robert Ozazuwa-Ness, a distinguished researcher at Microsoft Research, shares his unique expertise in this advanced guide, using a clear, code-driven approach to explain the fundamental details of causal machine learning that are often hidden in academic texts. Everything you learn in this course can be easily and effectively applied to industry challenges, from building interpretable causal models to predicting counterfactual outcomes.

What you will learn

Building Causal Reinforcement Learning algorithms.
Perform causal inference using modern probabilistic machine learning tools such as PyTorch and Pyro.
Compare and contrast statistical and econometric methods for causal inference.
Tuning algorithms for attribution, credit assignment, and explanation.
Transforming domain expertise into explainable causal models.
End-to-end causal inference with DoWhy.
Deep Bayesian Causal Generative AI Models.
A code-oriented overview of Do-Calculus and Pearl’s Causal Hierarchy.
Coding for Fine-Tuning Causal Large Language Models.
Integrating causal assumptions into deep learning architectures, including reinforcement learning and large language models.
Using PyTorch, Pyro, and other ML libraries to scale causal inference.