Year of Publication

2021

Degree Name

Master of Science (MS)

Document Type

Master's Thesis

College

Engineering

Department/School/Program

Computer Science

First Advisor

Dr. Brent Harrison

Abstract

We present Markov Decision Processes with Embedded Agents (MDPEAs), an extension of multi-agent POMDPs that allow for the modeling of environments that can change the actuators, sensors, and learning function of the agent, e.g., a household robot which could gain and lose hardware from its frame, or a sovereign software agent which could encounter viruses on computers that modify its code. We show several toy problems for which standard reinforcement-learning methods fail to converge, and give an algorithm, `just-copy-it`, which learns some of them. Unlike MDPs, MDPEAs are closed systems and hence their evolution over time can be treated as a Markov chain. In future work, we hope MDPEAs can be extended to model even fully embedded agents acting in real digital or physical environments.

Digital Object Identifier (DOI)

https://doi.org/10.13023/etd.2021.135

Share

COinS