Modern Deep Reinforcement Learning Algorithms
arXiv:1906.10025v243 citations
Originality Synthesis-oriented
AI Analysis
It provides a synthesis for researchers in AI and machine learning, but is incremental as it is a review paper.
The paper reviews the latest deep reinforcement learning algorithms, focusing on their theoretical justification, practical limitations, and empirical properties, without presenting new experimental results.
Recent advances in Reinforcement Learning, grounded on combining classical theoretical results with Deep Learning paradigm, led to breakthroughs in many artificial intelligence tasks and gave birth to Deep Reinforcement Learning (DRL) as a field of research. In this work latest DRL algorithms are reviewed with a focus on their theoretical justification, practical limitations and observed empirical properties.