دانلود کتاب هوش مصنوعی سببی

I wrote this book because I wanted a code-first approach to causal inference that seamlessly fit with modern deep learning. It didn’t make sense to me that deep learning was often presented as being at odds with causal reasoning and inference, so I wanted to write a book that proved they combine well to their mutual benefit. Second, I wanted to close an obvious gap. Deep generative machine learning methods and graphical causal inference have a common ancestor in probabilistic graphical models. There have been tremendous advances in generative machine learning in recent years, including in the ability to synthetize realistic text, images, and video. Yet, in my view, the low-hanging fruit of connections to related concepts in graphical causality was left to rot on the vine. Chances are that if you’re reading this, you sensed this gap as well. So here we are. This book evolved from the Causal AI workshop I run through Altdeep.ai, an educational company that runs workshops and community events devoted to advanced topics in modeling. Participants in this causal AI workshop have included data scientists, machine learning engineers, and product managers from Google, Amazon, Meta, and other big tech companies. They’ve also included data scientists and ML experts from retailers such as Nike, consultancies like Deloitte, and pharmaceuticals like AstraZeneca. We’ve worked with quantitative marketing experts trying to take causal approaches to channel attribution. We’ve worked with economists and molecular biologists trying to get a more general perspective on the causal methods popular in their domains. We’ve worked with professors, post-docs, and PhD students across departments looking for a code-first approach to learning causal inference. I wrote this book for all of these people, based on the realworld problems they care about and their feedback. If you belong to or relate to any of these groups, this book is for you, too. How is this book different from other causal inference books? Causal inference relies mainly on three different skill sets: the ability to turn your domain knowledge into a causal model rendered in code, deep skills in probability theory, and deep skills in statistical theory and methods. This book focuses on the first skill by using libraries that enable bespoke causal modeling, and by leveraging the deep learning machinery in tools such as PyTorch to do the statistical heavy lifting. I hope this sounds like what you are looking for.