Artificial Intelligence and Causal Inference

Artificial Intelligence and Causal Inference discuss the recent development of relationships between AI and causal inference,which is a key component for understanding intelligence. This book covers three types of neural networks,formulates deep learning as an optimal control problem,and uses Pontryagins Maximum Principle for network training. It also uses RL to infer large causal networks,VAE,GAN,neural differential equations,recurrent neural network (RNN),and RL to estimate counterfactual outcomes. AI-based methods for estimation of individualized treatment effect in the presence of network interference are also discussed.

Format: Hardback
Length: 368 pages
Publication date: 08 March 2022
Publisher: Taylor & Francis Ltd

Artificial Intelligence and Causal Inference delve into the evolving connections between artificial intelligence (AI) and causal inference, highlighting the ongoing challenges and opportunities in AI development. Despite substantial advancements in AI, a central obstacle remains the need to comprehend the mechanisms underlying intelligence, encompassing reasoning, planning, and imagination. Understanding, transfer, and generalization are fundamental principles that fuel intelligence. Causal inference, a crucial component in understanding, encompasses intervention, domain shift learning, temporal structure, and counterfactual thinking. However, these essential elements of causality are often overlooked in machine learning, leading to failures in deep learning. This book aims to bridge the gap between AI and modern causal analysis, fostering the advancement of AI and its applications. It serves as a valuable resource for graduate students and researchers in AI, data science, causal inference, statistics, genomics, bioinformatics, and precision medicine.

Key Features:

Comprehensive Coverage: The book explores three types of neural networks, formulates deep learning as an optimal control problem, and utilizes Pontryagins Maximum Principle for network training. It also delves into deep learning for nonlinear mediation and instrumental variable causal analysis, constructing causal networks as continuous optimization problems.

Advanced Techniques: The book employs transformer and attention to encode-decode graphics, utilizes RL for inferring large causal networks, and employs VAE, GAN, neural differential equations, and recurrent neural networks (RNNs) for various AI applications.

Applications in AI and Beyond: The book showcases the practical applications of AI and causal inference in diverse fields, including healthcare, finance, social sciences, and natural sciences. It provides insights into how AI can be used to address complex problems and make informed decisions.

Conclusion: Artificial Intelligence and Causal Inference offer a comprehensive and up-to-date exploration of the intersection of AI and causal inference. By bridging the gap between these two domains, the book empowers researchers and practitioners to unlock the full potential of AI and drive its transformative impact across industries.

Weight: 1160g
Dimension: 217 x 285 x 30 (mm)
ISBN-13: 9780367859404