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
Welfare and felt duration – Andreas Mogensen (Global Priorities Institute, University of Oxford)
How should we understand the duration of a pleasant or unpleasant sensation, insofar as its duration modulates how good or bad the experience is overall? Given that we seem able to distinguish between subjective and objective duration and that how well or badly someone’s life goes is naturally thought of as something to be assessed from her own perspective, it seems intuitive that it is subjective duration that modulates how good or bad an experience is from the perspective of an individual’s welfare. …
Evolutionary debunking and value alignment – Michael T. Dale (Hampden-Sydney College) and Bradford Saad (Global Priorities Institute, University of Oxford)
This paper examines the bearing of evolutionary debunking arguments—which use the evolutionary origins of values to challenge their epistemic credentials—on the alignment problem, i.e. the problem of ensuring that highly capable AI systems are properly aligned with values. Since evolutionary debunking arguments are among the best empirically-motivated arguments that recommend changes in values, it is unsurprising that they are relevant to the alignment problem. However, how evolutionary debunking arguments…
Tiny probabilities and the value of the far future – Petra Kosonen (Population Wellbeing Initiative, University of Texas at Austin)
Morally speaking, what matters the most is the far future – at least according to Longtermism. The reason why the far future is of utmost importance is that our acts’ expected influence on the value of the world is mainly determined by their consequences in the far future. The case for Longtermism is straightforward: Given the enormous number of people who might exist in the far future, even a tiny probability of affecting how the far future goes outweighs the importance of our acts’ consequences…