# Emergence and Causality in Complex Systems: A Survey on Causal Emergence and Related Quantitative Studies

**Authors:** Bing Yuan, Zhang Jiang, Aobo Lyu, Jiayun Wu, Zhipeng Wang, Mingzhe Yang, Kaiwei Liu, Muyun Mou, Peng Cui  
**arXiv:** [2312.16815](https://arxiv.org/abs/2312.16815)  
**Date:** 2023-12-28  
**Summary page:** https://www.summarizepaper.com/en/paper/2312.16815v1/

## Key points

- Emergence and causality are fundamental concepts in understanding complex systems
- Emergence refers to macroscopic properties that cannot be explained by individual properties alone
- Causality can exhibit emergence by giving rise to new causal laws as abstraction levels increase
- The theory of causal emergence aims to bridge these concepts and uses measures of causality to quantify emergence
- The paper titled "Emergence and Causality in Complex Systems: A Survey on Causal Emergence and Related Quantitative Studies" provides a comprehensive review of recent advancements in quantitative theories and applications related to causal emergence
- The authors address two key problems: quantifying causal emergence and identifying it in data
- Machine learning techniques are used to tackle the issue of identifying causal emergence in data, connecting it with artificial intelligence
- The architectures used for identifying causal emergence are shared by other areas such as causal representation learning, causal model abstraction, and world model-based reinforcement learning
- Quantifying causal emergence is important and can be measured using various approaches discussed in the paper
- Machine learning techniques are also discussed for identifying causal emergence in real-world data
- Potential applications of causal emergence theory are highlighted across different domains such as social sciences, economics, biology, physics, and computer science
- Future perspectives on further research in understanding complex systems through causality are provided

## Summary

Emergence and causality are two fundamental concepts that are interconnected and essential for understanding complex systems. Emergence refers to the phenomenon where macroscopic properties cannot be solely explained by individual properties, while causality can exhibit emergence by giving rise to new causal laws as we increase the level of abstraction. The theory of causal emergence aims to bridge these concepts and uses measures of causality to quantify emergence. In their paper titled "Emergence and Causality in Complex Systems: A Survey on Causal Emergence and Related Quantitative Studies," authors Bing Yuan, Zhang Jiang, Aobo Lyu, Jiayun Wu, Zhipeng Wang, Mingzhe Yang, Kaiwei Liu, Muyun Mou, and Peng Cui provide a comprehensive review of recent advancements in quantitative theories and applications related to causal emergence. They address two key problems: quantifying causal emergence and identifying it in data. To tackle the latter issue, they utilize machine learning techniques which establish a connection between causal emergence and artificial intelligence. The authors highlight that the architectures used for identifying causal emergence are shared by causal representation learning, causal model abstraction, and world model-based reinforcement learning. Therefore, progress in any of these areas can benefit the others. The paper emphasizes the importance of quantifying causal emergence and provides insights into how it can be measured using various approaches. Additionally, it discusses methods for identifying causal emergence in real-world data through machine learning techniques. Furthermore,the authors discuss potential applications of causal emergence theory across different domains such as social sciences,economics,biology ,physics,and computer science.They also provide future perspectives on how further research can advance our understanding of complex systems through the lens of causality.This paper serves as a comprehensive survey on recent developments in quantitative theoriesand applications related tocausal emergencethat highlights the interconnectedness between emergent phenomena and causality while emphasizing the role of machine learning techniques in identifying and quantifying causal emergence. The insights provided in this review have implications for various fields and pave the way for future research in understanding complex systems.