Video Prediction Models as Rewards for Reinforcement Learning

Specifying reward signals that allow agents to learn complex behaviors is a long-standing challenge in reinforcement learning. A promising approach is to extract preferences for behaviors from unlabeled videos, which are widely available on the internet. We present Video Prediction Rewards (VIPER),...

Ausführliche Beschreibung

Bibliographische Detailangaben

Veröffentlicht in:	arXiv.org 2023-05
Hauptverfasser:	Escontrela, Alejandro, Adeniji, Ademi, Wilson, Yan, Jain, Ajay, Xue Bin Peng, Goldberg, Ken, Lee, Youngwoon, Hafner, Danijar, Abbeel, Pieter
Format:	Artikel
Sprache:	eng
Schlagworte:	Algorithms Prediction models Source code Video Websites
Online-Zugang:	Volltext
Tags:	Tag hinzufügen Keine Tags, Fügen Sie den ersten Tag hinzu!