Multi-agent Online Planning Architecture for Real-time Compliance - Sorbonne Université Access content directly
Conference Papers Year : 2023

Multi-agent Online Planning Architecture for Real-time Compliance

Hisashi Hayashi
  • Function : Author
Theodoros Mitsikas
  • Function : Author
Kanae Tsushima
  • Function : Author
Ralph Schäfermeier
  • Function : Author
Adrian Paschke
  • Function : Author
Ken Satoh
  • Function : Author

Abstract

Due to the recent rapid introduction of AI technologies into society, we face new risks related to AI. Therefore, it is very important to let AI be compliant with legal and ethical norms to reduce such risks. In this paper, we propose a method of compliance mechanism of AI agent planning in a multi-agent setting. We formalize legal norms as hard constraints which must be satisfied and ethical norms as soft constraints which should be satisfied as much as possible. Moreover, our mechanism can handle real-time exogenous environmental change by a replanning mechanism. We believe that our architecture is an important step towards "trustworthy AI".
Fichier principal
Vignette du fichier
paper5801.pdf (1.57 Mo) Télécharger le fichier
Origin Publisher files allowed on an open archive

Dates and versions

hal-04320268 , version 1 (04-12-2023)

Identifiers

  • HAL Id : hal-04320268 , version 1

Cite

Hisashi Hayashi, Theodoros Mitsikas, Yousef Sojasi Taheri, Kanae Tsushima, Ralph Schäfermeier, et al.. Multi-agent Online Planning Architecture for Real-time Compliance. 17th International Rule Challenge and 7th Doctoral Consortium @ RuleML+RR 2023, Sep 2023, Oslo, Norway. ⟨hal-04320268⟩
34 View
8 Download

Share

Gmail Mastodon Facebook X LinkedIn More