5 papers
Generalised Bellman recurrence and three dualities in sequential decision-making
Fernando E. Rosas, David Hyland, Daniel Polani
What gives the Bellman equation its form? We show that the recursive properties of optimal value functions follow from three conditions: that the dynamics decomposes through suffic…
Safety from Honesty in a Disinterested AI Predictor
Yoshua Bengio, Oliver Richardson, Tomáš GavenÄiak +13
As AI systems become more capable, training procedures that optimize for downstream outcomes risk introducing implicit agency: goal-directed behavior that designers never specified…
From monoliths to modules: Decomposing transducers for efficient world modelling
Alexander Boyd, Franz Nowak, David Hyland +2
World models have been recently proposed as sandbox environments in which AI agents can be trained and evaluated before deployment. While realistic world models often have high com…
Learning POMDP World Models from Observations with Language-Model Priors
Valentin Six, Frederik Panse, Mathis Fajeau +7
Whether navigating a building, operating a robot, or playing a game, an agent that acts effectively in an environment must first learn an internal model of how that environment wor…
Possible Principles for Aligned Structure Learning Agents
Lancelot Da Costa, Tomáš GavenÄiak, David Hyland +5
This paper offers a roadmap for the development of scalable aligned artificial intelligence (AI) from first principle descriptions of natural intelligence. In brief, a possible pat…