Learning to Plan with Portable Symbols

Steven James, Benjamin Rosman, George Konidaris
July 2018
PDF Cite Project

Abstract

We present a framework for autonomously learning a portable symbolic representation that describes a collection of low-level continuous environments. We show that abstract representations can be learned in a task-independent space specific to the agent that, when combined with problem-specific information, can be used for planning. We demonstrate knowledge transfer in a video game domain where an agent learns portable, task-independent symbolic rules, and then learns instantiations of these rules on a per-task basis, reducing the number of samples required to learn a representation of a new task.

Type
Workshop paper
Publication
ICML/IJCAI/AAMAS 2018 Workshop on Planning and Learning
Symbolic Representations Reinforcement Learning Classical Planning Transfer Learning
Steven James
Steven James
Deputy Lab Director

My research interests include reinforcement learning and planning.

Benjamin Rosman
Benjamin Rosman
Lab Director

I am a Professor in the School of Computer Science and Applied Mathematics at the University of the Witwatersrand in Johannesburg. I work in robotics, artificial intelligence, decision theory and machine learning.