Towards shutdownable agents via stochastic choice
Elliott Thornley (Global Priorities Institute, University of Oxford), Alexander Roman (New College of Florida), Christos Ziakas (Independent), Leyton Ho (Brown University) and Louis Thomson (University of Oxford)
GPI Working Paper No. 16-2024
Some worry that advanced artificial agents may resist being shut down. The Incomplete Preferences Proposal (IPP) is an idea for ensuring that doesn’t happen. A key part of the IPP is using a novel ‘Discounted REward for Same-Length Trajectories (DREST)’ reward function to train agents to (1) pursue goals effectively conditional on each trajectory-length (be ‘USEFUL’), and (2) choose stochastically between different trajectory-lengths (be ‘NEUTRAL’ about trajectory-lengths). In this paper, we propose evaluation metrics for USEFULNESS and NEUTRALITY. We use a DREST reward function to train simple agents to navigate gridworlds, and we find that these agents learn to be USEFUL and NEUTRAL. Our results thus suggest that DREST reward functions could also train advanced agents to be USEFUL and NEUTRAL, and thereby make these advanced agents useful and shutdownable.
Other working papers
Dynamic public good provision under time preference heterogeneity – Philip Trammell (Global Priorities Institute and Department of Economics, University of Oxford)
I explore the implications of time preference heterogeneity for the private funding of public goods. The assumption that players use a common discount rate is knife-edge: relaxing it yields substantially different equilibria, for two reasons. First, time preference heterogeneity motivates intertemporal polarization, analogous to the polarization seen in a static public good game. In the simplest settings, more patient players spend nothing early in time and less patient players spending nothing later. Second…
Longtermist institutional reform – Tyler M. John (Rutgers University) and William MacAskill (Global Priorities Institute, Oxford University)
There is a vast number of people who will live in the centuries and millennia to come. Even if homo sapiens survives merely as long as a typical species, we have hundreds of thousands of years ahead of us. And our future potential could be much greater than that again: it will be hundreds of millions of years until the Earth is sterilized by the expansion of the Sun, and many trillions of years before the last stars die out. …
Concepts of existential catastrophe – Hilary Greaves (University of Oxford)
The notion of existential catastrophe is increasingly appealed to in discussion of risk management around emerging technologies, but it is not completely clear what this notion amounts to. Here, I provide an opinionated survey of the space of plausibly useful definitions of existential catastrophe. Inter alia, I discuss: whether to define existential catastrophe in ex post or ex ante terms, whether an ex ante definition should be in terms of loss of expected value or loss of potential…