Bookmark Saya

arXiv Open Access 2022

Opportunities and Challenges from Using Animal Videos in Reinforcement Learning for Navigation

Vittorio Giammarino James Queeney Lucas C. Carstensen Michael E. Hasselmo Ioannis Ch. Paschalidis
Lihat Sumber

Abstrak

We investigate the use of animal videos (observations) to improve Reinforcement Learning (RL) efficiency and performance in navigation tasks with sparse rewards. Motivated by theoretical considerations, we make use of weighted policy optimization for off-policy RL and describe the main challenges when learning from animal videos. We propose solutions and test our ideas on a series of 2D navigation tasks. We show how our methods can leverage animal videos to improve performance over RL algorithms that do not leverage such observations.

Topik & Kata Kunci

eess.SY cs.AI cs.CV cs.LG

Penulis (5)

V

Vittorio Giammarino

J

James Queeney

L

Lucas C. Carstensen

M

Michael E. Hasselmo

I

Ioannis Ch. Paschalidis

Format Sitasi

APA MLA BibTeX
Giammarino, V., Queeney, J., Carstensen, L.C., Hasselmo, M.E., Paschalidis, I.C. (2022). Opportunities and Challenges from Using Animal Videos in Reinforcement Learning for Navigation. https://arxiv.org/abs/2209.12347

Akses Cepat

Lihat di Sumber
Informasi Jurnal
Tahun Terbit
2022
Bahasa
en
Sumber Database
arXiv
Akses
Open Access ✓