Toward Complex and Structured Goals in Reinforcement Learning

Davidson, G., & Gureckis, T. M. (2024). Toward Complex and Structured Goals in Reinforcement Learning. Reinforcement Learning Conference (RLC) Finding the Frame Workshop.


Abstract

Goals play a central role in the study of agentic behavior. But what is a goal, and how should we best represent them? The traditional reinforcement learning answer is that all goals are expressible as the maximization of future rewards. While parsimonious, such a definition seems insufficient when viewed from both the perspective of humans specifying goals to machines and autotelic agents that self-propose tasks to explore and learn. We offer a critical perspective on the distillation of all goals directly into reward functions. We identify key features we believe goal representations ought to have, and then offer a proposal we believe meets those considerations.


Bibtex entry:

@inproceedings{davidson2024goalsframe,
	abstract = {Goals play a central role in the study of agentic behavior. But what is a goal, and how should we best represent them? The traditional reinforcement learning answer is that all goals are expressible as the maximization of future rewards. While parsimonious, such a definition seems insufficient when viewed from both the perspective of humans specifying goals to machines and autotelic agents that self-propose tasks to explore and learn. We offer a critical perspective on the distillation of all goals directly into reward functions. We identify key features we believe goal representations ought to have, and then offer a proposal we believe meets those considerations.},
	author = {Davidson, G. and Gureckis, T.M.},
	booktitle = {Reinforcement Learning Conference (RLC) Finding the Frame Workshop},
	keywords = {RL, Creativity, Games, ArtificialIntelligence, Goals},
	title = {Toward Complex and Structured Goals in Reinforcement Learning},
	year = {2024}}


QR Code:


Download SVG