Conference site ยป Proceedings

PyCID: A Python Library for Causal Influence Diagrams

James Fox
University of Oxford

Tom Everitt
DeepMind

Ryan Carey
University of Oxford

Eric Langlois
University of Toronto

Alessandro Abate
University of Oxford

Michael Wooldridge
University of Oxford

Abstract

Why did a decision maker select a certain decision? What behaviour does a certain objective incentivise? How can we improve this behaviour and ensure that a decision-maker chooses decisions with safer or fairer consequences? This paper introduces the Python package PyCID, built upon pgmpy, that implements (causal) influence diagrams, a widely used graphical modelling framework for decision-making problems. By providing a range of methods to solve and analyse (causal) influence diagrams, PyCID helps answer questions about behaviour and incentives in both single-agent and multi-agent settings.

Keywords

Influence Diagrams, Causal Models, Probabilistic Graphical Models, Game Theory, Decision Theory

DOI

10.25080/majora-1b6fd038-008

Bibtex entry

Full text PDF