Rainbow: Combining Improvements in Deep Reinforcement Learning
Proceedings of the AAAI Conference on Artificial Intelligence2018Vol. 32(1)
Citations Over TimeTop 1% of 2018 papers
Matteo Hessel, Joseph Modayil, Hado van Hasselt, Tom Schaul, Georg Ostrovski, Will Dabney, Dan Horgan, Bilal Piot, Mohammad Gheshlaghi Azar, David Silver
Abstract
The deep reinforcement learning community has made several independent improvements to the DQN algorithm. However, it is unclear which of these extensions are complementary and can be fruitfully combined. This paper examines six extensions to the DQN algorithm and empirically studies their combination. Our experiments show that the combination provides state-of-the-art performance on the Atari 2600 benchmark, both in terms of data efficiency and final performance. We also provide results from a detailed ablation study that shows the contribution of each component to overall performance.
Related Papers
- → A Comprehensive Analysis of Machine Learning Techniques in Biomedical Image Processing Using Convolutional Neural Network(2022)18 cited
- → Efficient Disease Risk Prediction based on Deep Learning Approach(2022)10 cited
- → A Survey on Deep Learning Models Embed Bio-Inspired Algorithms in Cardiac Disease Classification(2022)4 cited
- → Does Deep Learning REALLY Outperform Non-deep Machine Learning for Clinical Prediction on Physiological Time Series?(2022)2 cited
- → Machine Learning and Deep Learning applications-a vision using the SPSS Method(2020)