Value Inference in Sociotechnical Systems

Adaptive Agents and Multi-Agents Systems, pp. 1774–1780

Abstract

As artificial agents become increasingly embedded in our society, we must ensure that their behavior aligns with human values. Value alignment entails value inference, the process of identifying values and reasoning about how humans prioritize values. We introduce a holistic framework that connects the technical (AI) components necessary for value inference. Subsequently, we discuss how hybrid intelligence-the synergy of human and artificial intelligence---is instrumental to the success of value inference. Finally, we illustrate how value inference both poses significant challenges and provides novel opportunities for multiagent systems research.

Authors 8

1. Enrico Liscio

   Delft University of Technology

   Affiliation as printed

   TU Delft, Delft, Netherlands

2. Roger X. Lera-Leri

   Consejo Superior de Investigaciones Científicas · Artificial Intelligence Research Institute

   Affiliation as printed

   IIIA-CSIC, Barcelona, Spain

3. Filippo Bistaffa

   Consejo Superior de Investigaciones Científicas · Artificial Intelligence Research Institute

   Affiliation as printed

   IIIA-CSIC, Barcelona, Spain

4. Roel Dobbe

   Delft University of Technology

   Affiliation as printed

   TU Delft, Delft, Netherlands

5. Catholijn M. Jonker Aachen

   Leiden University · Delft University of Technology

   Affiliation as printed

   TU Delft & Leiden University, Delft, Netherlands

6. Maite Lopez-Sanchez

   Universitat de Barcelona

   Affiliation as printed

   University of Barcelona, Barcelona, Spain

7. Juan A. Rodríguez-Aguilar

   Consejo Superior de Investigaciones Científicas · Artificial Intelligence Research Institute

   Affiliation as printed

   IIIA-CSIC, Barcelona, Spain

8. Pradeep Kumar Murukannaiah

   Delft University of Technology

   Affiliation as printed

   TU Delft, Delft, Netherlands