Learning a World Model With Multitimescale Memory Augmentation

Model-based reinforcement learning (RL) is regarded as a promising approach to tackle the challenges that hinder model-free RL. The success of model-based RL hinges critically on the quality of the predicted dynamic models. However, for many real-world tasks involving high-dimensional state spaces,...

Ausführliche Beschreibung

Gespeichert in:

Bibliographische Detailangaben
Veröffentlicht in:	IEEE transaction on neural networks and learning systems 2023-11, Vol.34 (11), p.8493-8502
Hauptverfasser:	Cai, Wenzhe, Wang, Teng, Wang, Jiawei, Sun, Changyin
Format:	Artikel
Sprache:	eng
Schlagworte:	Adaptive optics Dynamic models Integrated optics Learning Long term memory Memory modules Model-based exploration multitimescale memory augmentation Neural networks Optical flow (image analysis) Optical imaging Prediction models Predictive models Recurrent neural networks reinforcement learning (RL) Short term memory Task analysis Training world model
Online-Zugang:	Volltext bestellen
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!

Schreiben Sie den ersten Kommentar!