Discovering agents
From MaRDI portal
Publication:6136104
Abstract: Causal models of agents have been used to analyse the safety aspects of machine learning systems. But identifying agents is non-trivial -- often the causal model is just assumed by the modeler without much justification -- and modelling failures can lead to mistakes in the safety analysis. This paper proposes the first formal causal definition of agents -- roughly that agents are systems that would adapt their policy if their actions influenced the world in a different way. From this we derive the first causal discovery algorithm for discovering agents from empirical data, and give algorithms for translating between causal models and game-theoretic influence diagrams. We demonstrate our approach by resolving some previous confusions caused by incorrect causal modelling of agents.
Cites work
- scientific article; zbMATH DE number 3118779 (Why is no real title available?)
- scientific article; zbMATH DE number 3159869 (Why is no real title available?)
- scientific article; zbMATH DE number 1467490 (Why is no real title available?)
- Actual causation and the art of modeling
- Bayesian Networks and Influence Diagrams
- Causal Inference Using the Algorithmic Markov Condition
- Causality for Machine Learning
- Causality. Models, reasoning, and inference
- Elements of causal inference. Foundations and learning algorithms
- Foundations of structural causal models with cycles and latent variables
- Graphs for margins of Bayesian networks
- Influence Diagrams for Causal Modelling and Inference
- Information-geometric approach to inferring causal directions
- Multi-agent influence diagrams for representing and solving games.
- Reasoning about causality in games
- Reinforcement learning. An introduction
- Settable systems: an extension of Pearl's causal model with optimization, equilibrium, and learning
This page was built for publication: Discovering agents
Report a bug (only for logged in users!)Click here to report a bug for this page (MaRDI item Q6136104)
