AI Alignment: A Comprehensive Survey

AI alignment aims to make AI systems behave in line with human intentions and values. As AI systems grow more capable, so do risks from misalignment. To provide a comprehensive and up-to-date overview of the alignment field, in this survey, we delve into the core concepts, methodology, and practice...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:arXiv.org 2024-05
Hauptverfasser: Ji, Jiaming, Qiu, Tianyi, Chen, Boyuan, Zhang, Borong, Hantao Lou, Wang, Kaile, Duan, Yawen, He, Zhonghao, Zhou, Jiayi, Zhang, Zhaowei, Zeng, Fanzhi, Ng, Kwan Yee, Dai, Juntao, Pan, Xuehai, O'Gara, Aidan, Yingshan Lei, Xu, Hua, Tse, Brian, Fu, Jie, McAleer, Stephen, Yang, Yaodong, Wang, Yizhou, Song-Chun, Zhu, Guo, Yike, Gao, Wen
Format: Artikel
Sprache:eng
Schlagworte:
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:AI alignment aims to make AI systems behave in line with human intentions and values. As AI systems grow more capable, so do risks from misalignment. To provide a comprehensive and up-to-date overview of the alignment field, in this survey, we delve into the core concepts, methodology, and practice of alignment. First, we identify four principles as the key objectives of AI alignment: Robustness, Interpretability, Controllability, and Ethicality (RICE). Guided by these four principles, we outline the landscape of current alignment research and decompose them into two key components: forward alignment and backward alignment. The former aims to make AI systems aligned via alignment training, while the latter aims to gain evidence about the systems' alignment and govern them appropriately to avoid exacerbating misalignment risks. On forward alignment, we discuss techniques for learning from feedback and learning under distribution shift. On backward alignment, we discuss assurance techniques and governance practices. We also release and continually update the website (www.alignmentsurvey.com) which features tutorials, collections of papers, blog posts, and other resources.
ISSN:2331-8422