Corpus ID: 52172123

ARCHER: Aggressive Rewards to Counter bias in Hindsight Experience Replay

@article{Lanka2018ARCHERAR,
  title={ARCHER: Aggressive Rewards to Counter bias in Hindsight Experience Replay},
  author={S. Lanka and Tianfu Wu},
  journal={ArXiv},
  year={2018},
  volume={abs/1809.02070}
}
Experience replay is an important technique for addressing sample-inefficiency in deep reinforcement learning (RL), but faces difficulty in learning from binary and sparse rewards due to disproportionately few successful experiences in the replay buffer. Hindsight experience replay (HER) was recently proposed to tackle this difficulty by manipulating unsuccessful transitions, but in doing so, HER introduces a significant bias in the replay buffer experiences and therefore achieves a suboptimal… Expand
13 Citations
ACDER: Augmented Curiosity-Driven Experience Replay
  • PDF
Trial and Error Experience Replay Based Deep Reinforcement Learning
  • C. Zhang, Liang Ma
  • Computer Science
  • 2019 IEEE International Conference on Smart Cloud (SmartCloud)
  • 2019
  • 1
Reinforcement Learning with Converging Goal Space and Binary Reward Function
Counterfactual Data Augmentation using Locally Factored Dynamics
  • 3
  • PDF
Universal Value Density Estimation for Imitation Learning and Goal-Conditioned Reinforcement Learning
  • 2
  • PDF
Hindsight Generative Adversarial Imitation Learning
  • 2
  • PDF
...
1
2
...

References

SHOWING 1-10 OF 52 REFERENCES
Hindsight Experience Replay
  • 747
  • PDF
A Deeper Look at Experience Replay
  • 86
  • PDF
Curiosity-Driven Exploration by Self-Supervised Prediction
  • 940
  • PDF
Prioritized Experience Replay
  • 1,533
  • PDF
Learning by Playing - Solving Sparse Reward Tasks from Scratch
  • 191
  • PDF
Hierarchical Deep Reinforcement Learning: Integrating Temporal Abstraction and Intrinsic Motivation
  • 614
  • PDF
Human-level control through deep reinforcement learning
  • 11,642
  • PDF
TEXPLORE: real-time sample-efficient reinforcement learning for robots
  • 92
  • PDF
...
1
2
3
4
5
...