Cargando…
Prioritized experience replay in path planning via multi-dimensional transition priority fusion
INTRODUCTION: Deep deterministic policy gradient (DDPG)-based path planning algorithms for intelligent robots struggle to discern the value of experience transitions during training due to their reliance on a random experience replay. This can lead to inappropriate sampling of experience transitions...
Autores principales: | Cheng, Nuo, Wang, Peng, Zhang, Guangyuan, Ni, Cui, Nematov, Erkin |
---|---|
Formato: | Online Artículo Texto |
Lenguaje: | English |
Publicado: |
Frontiers Media S.A.
2023
|
Materias: | |
Acceso en línea: | https://www.ncbi.nlm.nih.gov/pmc/articles/PMC10684733/ https://www.ncbi.nlm.nih.gov/pubmed/38034838 http://dx.doi.org/10.3389/fnbot.2023.1281166 |
Ejemplares similares
-
Prioritized memory access explains planning and hippocampal replay
por: Mattar, Marcelo G., et al.
Publicado: (2018) -
The roles of online and offline replay in planning
por: Eldar, Eran, et al.
Publicado: (2020) -
Offline replay supports planning in human reinforcement learning
por: Momennejad, Ida, et al.
Publicado: (2018) -
Heuristic Q-learning based on experience replay for three-dimensional path planning of the unmanned aerial vehicle
por: Xie, Ronglei, et al.
Publicado: (2019) -
Let’s replay
por: Rasch, Björn
Publicado: (2018)