Corpus ID: 54457634

Learning to Design Circuits

@article{Wang2018LearningTD,
  title={Learning to Design Circuits},
  author={Hanrui Wang and J. Yang and H. Lee and Song Han},
  journal={ArXiv},
  year={2018},
  volume={abs/1812.02734}
}
  • Hanrui Wang, J. Yang, +1 author Song Han
  • Published 2018
  • Computer Science, Mathematics
  • ArXiv
  • Analog IC design relies on human experts to search for parameters that satisfy circuit specifications with their experience and intuitions, which is highly labor intensive, time consuming and suboptimal. Machine learning is a promising tool to automate this process. However, supervised learning is difficult for this task due to the low availability of training data: 1) Circuit simulation is slow, thus generating large-scale dataset is time-consuming; 2) Most circuit designs are propitiatory IPs… CONTINUE READING
    AutoCkt: Deep Reinforcement Learning of Analog Circuit Designs
    • 3
    • PDF
    GCN-RL Circuit Designer: Transferable Transistor Sizing with Graph Neural Networks and Reinforcement Learning
    • 3
    • PDF
    Exploring a Machine Learning Approach to Performance Driven Analog IC Placement
    • 1
    • PDF
    HAT: Hardware-Aware Transformers for Efficient Natural Language Processing
    • 8
    • PDF
    Reinforcement Learning Driven Physical Synthesis

    References

    Publications referenced by this paper.
    SHOWING 1-10 OF 20 REFERENCES
    An Efficient Bayesian Optimization Approach for Automated Optimization of Analog Circuits
    • 29
    • PDF
    Sequence to Sequence Learning with Neural Networks
    • 10,529
    • PDF
    Design of Analog CMOS Integrated Circuits
    • 2,449
    • PDF
    Design Of Analog Cmos Integrated Circuits
    • 579
    • PDF
    Parameter Space Noise for Exploration
    • 271
    • PDF
    Continuous control with deep reinforcement learning
    • 3,529
    • Highly Influential
    • PDF