Scalar Reward is Not Enough

Peter Vamplew (Federation University Australia), Benjamin J. Smith (University of Oregon), Johan Källström (Linköping University), Gabriel Ramos (Universidade do Vale do Rio dos Sinos), Roxana Rădulescu (Vrije Universiteit Brussel), Diederik M. Roijers (Vrije Universiteit Brussel), Conor F. Hayes (University of Galway), Friedrik Hentz (Linköping University), Patrick Mannion (University of Galway), Pieter J.K. Libin (Vrije Universiteit Brussel), Richard Dazeley (Deakin University), Cameron Foale (Federation University Australia)

Abstract

Silver et al. [14] posit that scalar reward maximisation is sufficient to underpin all intelligence and provides a suitable basis for artificial general intelligence (AGI). This extended abstract summarises the counter-argument from our JAAMAS paper[19].