Deep Ensemble Reinforcement Learning with Multiple Deep Deterministic Policy Gradient Algorithm

Deep deterministic policy gradient algorithm operating over continuous space of actions has attracted great attention for reinforcement learning. However, the exploration strategy through dynamic programming within the Bayesian belief state space is rather inefficient even for simple systems. Anothe...

Ausführliche Beschreibung

Gespeichert in:
Bibliographische Detailangaben
Veröffentlicht in:Mathematical problems in engineering Jg. 2020; H. 2020; S. 1 - 12
Hauptverfasser: Wu, Junta, Li, Huiyun
Format: Journal Article
Sprache:Englisch
Veröffentlicht: Cairo, Egypt Hindawi Publishing Corporation 2020
Hindawi
John Wiley & Sons, Inc
Schlagworte:
Algorithms
Computer simulation
Dynamic programming
Efficiency
Interactive learning
Machine learning
Mathematical problems
Methods
Neural networks
Race cars
Robot arms
Statistical methods
Training
ISSN:1024-123X, 1563-5147
Online-Zugang:Volltext
Tags: Tag hinzufügen
Keine Tags, Fügen Sie den ersten Tag hinzu!
Beschreibung
Zusammenfassung:Deep deterministic policy gradient algorithm operating over continuous space of actions has attracted great attention for reinforcement learning. However, the exploration strategy through dynamic programming within the Bayesian belief state space is rather inefficient even for simple systems. Another problem is the sequential and iterative training data with autonomous vehicles subject to the law of causality, which is against the i.i.d. (independent identically distributed) data assumption of the training samples. This usually results in failure of the standard bootstrap when learning an optimal policy. In this paper, we propose a framework of m-out-of-n bootstrapped and aggregated multiple deep deterministic policy gradient to accelerate the training process and increase the performance. Experiment results on the 2D robot arm game show that the reward gained by the aggregated policy is 10%–50% better than those gained by subpolicies. Experiment results on the open racing car simulator (TORCS) demonstrate that the new algorithm can learn successful control policies with less training time by 56.7%. Analysis on convergence is also given from the perspective of probability and statistics. These results verify that the proposed method outperforms the existing algorithms in both efficiency and performance.
Bibliographie:ObjectType-Article-1
SourceType-Scholarly Journals-1
ObjectType-Feature-2
content type line 14
ISSN:1024-123X
1563-5147
DOI:10.1155/2020/4275623