Combining Imagination and Heuristics to Learn Strategies that Generalize
Neurons, Behavior, Data analysis, and Theory · 2020
DOI: 10.51628/001c.13477 ↗Kurz gefasst
Fachartikel in der Zeitschrift Neurons, Behavior, Data analysis, and Theory, 2020. Verfasst von Erik Peterson und 3 weiteren Autor:innen.
Abstract (Originalsprache)
Deep reinforcement learning can match or exceed human performance in stable contexts, but with minor changes to the environment artificial networks, unlike humans, often cannot adapt. Humans rely on a combination of heuristics to simplify computational load and imagination to extend experiential learning to new and more challenging environments. Motivated by theories of the hierarchical organization of the human prefrontal networks, we have developed a model of hierarchical reinforcement learning that combines both heuristics and imagination into a “stumbler-strategist” network. We test performance of this network using Wythoff’s game, a gridworld environment with a known optimal strategy. We show that a heuristic labeling of each position as hot or cold, combined with imagined play, both accelerates learning and promotes transfer to novel games, while also improving model interpretability
Zitieren
APA
Peterson, E., Müyesser, N. A., Verstynen, T., & Dunovan, K. (2020). Combining Imagination and Heuristics to Learn Strategies that Generalize. Neurons, Behavior, Data analysis, and Theory. https://doi.org/10.51628/001c.13477
Vancouver
Peterson E, Müyesser NA, Verstynen T, Dunovan K. Combining Imagination and Heuristics to Learn Strategies that Generalize. Neurons, Behavior, Data analysis, and Theory. 2020. doi:10.51628/001c.13477
BibTeX
@article{Peterson2020,
author = {Peterson, Erik and Müyesser, Necati Alp and Verstynen, Timothy and Dunovan, Kyle},
title = {Combining Imagination and Heuristics to Learn Strategies that Generalize},
journal = {Neurons, Behavior, Data analysis, and Theory},
year = {2020},
doi = {10.51628/001c.13477}
}
EUVIMED