Corpus ID: 3522594

Structured Control Nets for Deep Reinforcement Learning

@article{Srouji2018StructuredCN,
  title={Structured Control Nets for Deep Reinforcement Learning},
  author={Mario Srouji and Jian Zhang and R. Salakhutdinov},
  journal={ArXiv},
  year={2018},
  volume={abs/1802.08311}
}
  • Mario Srouji, Jian Zhang, R. Salakhutdinov
  • Published 2018
  • Computer Science, Mathematics
  • ArXiv
  • In recent years, Deep Reinforcement Learning has made impressive advances in solving several important benchmark problems for sequential decision making. Many control applications use a generic multilayer perceptron (MLP) for non-vision parts of the policy network. In this work, we propose a new neural network architecture for the policy network representation that is simple yet effective. The proposed Structured Control Net (SCN) splits the generic MLP into two separate sub-modules: a… CONTINUE READING
    27 Citations
    Integrating Classical Control into Reinforcement Learning Policy
    • Highly Influenced
    Regularization Matters in Policy Optimization
    • 15
    Regularization Matters in Policy Optimization -- An Empirical Study on Continuous Control.
    • 1
    • PDF
    Recurrent Control Nets as Central Pattern Generators for Deep Reinforcement Learning
    • Highly Influenced
    Investigating Generalisation in Continuous Deep Reinforcement Learning
    • 21
    • PDF
    Generalization in Transfer Learning
    • 2
    • PDF
    Generalization in Reinforcement Learning with Selective Noise Injection and Information Bottleneck
    • 27
    • PDF
    Residual Reinforcement Learning for Robot Control
    • 74
    • PDF
    Optimal Control of Blank Holder Force Based on Deep Reinforcement Learning
    • P. Guo, Jianbo Yu
    • Computer Science
    • 2019 IEEE International Conference on Industrial Engineering and Engineering Management (IEEM)
    • 2019

    References

    SHOWING 1-10 OF 33 REFERENCES
    Continuous control with deep reinforcement learning
    • 4,301
    • PDF
    Dueling Network Architectures for Deep Reinforcement Learning
    • 1,380
    • Highly Influential
    • PDF
    End-to-End Training of Deep Visuomotor Policies
    • 1,984
    • PDF
    Asynchronous Methods for Deep Reinforcement Learning
    • 3,780
    • PDF
    Human-level control through deep reinforcement learning
    • 11,326
    • PDF
    Value Prediction Network
    • 177
    • PDF
    Least-Squares Temporal Difference Learning for the Linear Quadratic Regulator
    • 78
    • PDF
    Deep Reinforcement Learning that Matters
    • 748
    • Highly Influential
    • PDF