Corpus ID: 155092294

Distributional Reinforcement Learning for Efficient Exploration

@article{Mavrin2019DistributionalRL,
  title={Distributional Reinforcement Learning for Efficient Exploration},
  author={B. Mavrin and S. Zhang and Hengshuai Yao and Linglong Kong and Kaiwen Wu and Y. Yu},
  journal={ArXiv},
  year={2019},
  volume={abs/1905.06125}
}
In distributional reinforcement learning (RL), the estimated distribution of value function models both the parametric and intrinsic uncertainties. We propose a novel and efficient exploration method for deep RL that has two components. The first is a decaying schedule to suppress the intrinsic uncertainty. The second is an exploration bonus calculated from the upper quantiles of the learned distribution. In Atari 2600 games, our method outperforms QR-DQN in 12 out of 14 hard games (achieving… Expand
16 Citations
Non-crossing quantile regression for deep reinforcement learning
  • Highly Influenced
  • PDF
Bayesian Distributional Policy Gradients
  • Luchen Li, A. Faisal
  • Computer Science
  • ArXiv
  • 2021
  • PDF
Amortized Variational Deep Q Network
  • PDF
Data Efficient Training for Reinforcement Learning with Adaptive Behavior Policy Sharing
  • 1
  • PDF
Being Optimistic to Be Conservative: Quickly Learning a CVaR Policy
  • 10
  • PDF
Deep Reinforcement Learning with Decorrelation
  • 5
  • PDF
Efficient exploration of zero-sum stochastic games
  • 1
  • PDF
A Distributional Perspective on Value Function FactorizationMethods for Multi-Agent Reinforcement Learning
  • Wei-Fang Sun, Cheng-Kuang Lee, Chun-Yi Lee
  • 2021
  • PDF
...
1
2
...

References

SHOWING 1-10 OF 45 REFERENCES
Exploration in the Face of Parametric and Intrinsic Uncertainties
  • 1
  • PDF
Distributional Reinforcement Learning with Quantile Regression
  • 171
  • Highly Influential
  • PDF
QUOTA: The Quantile Option Architecture for Reinforcement Learning
  • 10
  • PDF
A Distributional Perspective on Reinforcement Learning
  • 523
  • Highly Influential
  • PDF
Distributed Distributional Deterministic Policy Gradients
  • 178
  • PDF
#Exploration: A Study of Count-Based Exploration for Deep Reinforcement Learning
  • 341
  • PDF
Parametric Return Density Estimation for Reinforcement Learning
  • 56
  • PDF
The Uncertainty Bellman Equation and Exploration
  • 77
  • PDF
An Analysis of Categorical Distributional Reinforcement Learning
  • 38
  • PDF
...
1
2
3
4
5
...