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 does not 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
Cassandra’s Curse: A second tragedy of the commons – Philippe Colo (ETH Zurich)
This paper studies why scientific forecasts regarding exceptional or rare events generally fail to trigger adequate public response. I consider a game of contribution to a public bad. Prior to the game, I assume contributors receive non-verifiable expert advice regarding uncertain damages. In addition, I assume that the expert cares only about social welfare. Under mild assumptions, I show that no information transmission can happen at equilibrium when the number of contributors…
The unexpected value of the future – Hayden Wilkinson (Global Priorities Institute, University of Oxford)
Various philosophers accept moral views that are impartial, additive, and risk-neutral with respect to betterness. But, if that risk neutrality is spelt out according to expected value theory alone, such views face a dire reductio ad absurdum. If the expected sum of value in humanity’s future is undefined—if, e.g., the probability distribution over possible values of the future resembles the Pasadena game, or a Cauchy distribution—then those views say that no real-world option is ever better than any other. And, as I argue…
Time discounting, consistency and special obligations: a defence of Robust Temporalism – Harry R. Lloyd (Yale University)
This paper defends the claim that mere temporal proximity always and without exception strengthens certain moral duties, including the duty to save – call this view Robust Temporalism. Although almost all other moral philosophers dismiss Robust Temporalism out of hand, I argue that it is prima facie intuitively plausible, and that it is analogous to a view about special obligations that many philosophers already accept…