Weighted Double Q-learning

By Zongzhang Zhang, Zhiyuan Pan, Mykel J. Kochenderfer. 2017

121 citations recorded by the source corpus.

Subfields: Artificial Intelligence, Computational Theory and Mathematics

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Hongliang Guo
2. Min He
3. Yangchen Pan
4. Yan Zheng
5. Jianye Hao
6. Keith W. Ross
7. Che Wang
8. Zijian Zhou

In Computational Theory and Mathematics

1. Hongliang Guo
2. Min He
3. Yan Zheng
4. Keith W. Ross
5. Zijian Zhou
6. Jianye Hao
7. Che Wang
8. Martha White

Citations in the corpus

• Controlling Overestimation Bias with Truncated Mixture of Continuous Distributional Quantile Critics (2020, neutral): “Subse-quently, (Zhang et al., 2017) and Lv et al. (2019) introduced the Weighted Estimator in the reinforcement learning community.”
• Interleaved Q-Learning with Partially Coupled Training Process (2019, methodology): “7It is worth noting that these two types of scenarios are also selected as benchmark scenarios for the evaluation of double Q-learning [8], and weighted double Q-learning [26].”
• Randomized Ensembled Double Q-Learning: Learning Fast Without a Model (2021, methodology): “Other bias reduction techniques include using bias-correction terms (Lee et al., 2013), using weighted Q estimates (Zhang et al., 2017; Li & Hou, 2019), penalizing deterministic policies at early stage of training (Fox et al., 2015), using multi-step methods (Meng et al., 2020), performing weighted…”
• A Q-values Sharing Framework for Multi-agent Reinforcement Learning under Budget Constraint (2020, background): “Furthermore, PSAF can be linked to many learning algorithm, such as double Q-learning [41], weighted double Q-learning [42], and an extension Multi Q-Learning [43].”
• Maxmin Q-learning: Controlling the Estimation Bias of Q-learning (2020, methodology): “Note that Maxmin Q-learning uses a different mechanism to reduce overestimation bias than Double Q-learning; Maxmin Q-learning with N = 2 is not Double Q-learning.”
• Weighted Double Deep Multiagent Reinforcement Learning in Stochastic Cooperative Environments (2018, methodology): “First, we extend weighted double Q-learning (WDQ) [Zhang et al. , 2017], a state-of-the-art traditional RL method, to the multiagent DRL settings.”