Emergence of Agency from a Causal Perspective
Typ
Examensarbete för masterexamen
Master's Thesis
Master's Thesis
Program
Data science and AI (MPDSC), MSc
Publicerad
2024
Författare
Ånestrand, Alvin
Modellbyggare
Tidskriftstitel
ISSN
Volymtitel
Utgivare
Sammanfattning
Causal models of agents and agentic behavior allows for safety analysis of machine
learning systems. Understanding how goal-directed behavior emerges from adapting
to an environment is however non-trivial. This thesis addresses the gap between
theoretical models and real-world implementations of machine learning systems,
though a framework that formalizes the connection between system dynamics and
goal-driven behavior.
This thesis introduces novel probabilistic graphical models for describing system
dynamics involving learning agents, based on dynamic bayesian networks, which
allows for a flexible representation of causal relationships in the training environment.
To analyze goal-directed behavior that emerges from interacions between agents and
the environment, the thesis also introduces temporally abstracted models. Such a
model captures the dynamics of a system after the learning process has converged,
derived from a model of the learning process. A temporally abstracted model describes
potential outcomes involving equilibria between agents and the environment, and
can under certain conditions be viewed as a model of goal-directed behavior in the
system.
Beskrivning
Ämne/nyckelord
Intelligent agent, objective, causal influence diagram, causal model, emergence, learning, equilibria, dynamical systems, goal-directed behavior