Corpus ID: 225061993

Bridging the Gap between Sample-based and One-shot Neural Architecture Search with BONAS

@article{Shi2020BridgingTG,
  title={Bridging the Gap between Sample-based and One-shot Neural Architecture Search with BONAS},
  author={Han Shi and R. Pi and Hang Xu and Zhenguo Li and James T. Kwok and Tong Zhang},
  journal={arXiv: Learning},
  year={2020}
}
  • Han Shi, R. Pi, +3 authors Tong Zhang
  • Published 2020
  • Computer Science, Mathematics
  • arXiv: Learning
  • Neural Architecture Search (NAS) has shown great potentials in finding better neural network designs. Sample-based NAS is the most reliable approach which aims at exploring the search space and evaluating the most promising architectures. However, it is computationally very costly. As a remedy, the one-shot approach has emerged as a popular technique for accelerating NAS using weight-sharing. However, due to the weight-sharing of vastly different networks, the one-shot approach is less reliable… CONTINUE READING
    7 Citations

    Figures and Tables from this paper

    Fitting the Search Space of Weight-sharing NAS with Graph Convolutional Networks
    • 3
    • PDF
    TRANSNAS-BENCH-101: IMPROVING TRANSFERRA-
    • 2020
    • PDF
    VEGA: Towards an End-to-End Configurable AutoML Pipeline
    • 1
    • PDF
    CATE: Computation-aware Neural Architecture Encoding with Transformers
    • Highly Influenced
    • PDF
    Automated Machine Learning on Graphs: A Survey
    • PDF

    References

    SHOWING 1-10 OF 45 REFERENCES
    Adaptive Stochastic Natural Gradient Method for One-Shot Neural Architecture Search
    • 30
    • PDF
    BayesNAS: A Bayesian Approach for Neural Architecture Search
    • 74
    • PDF
    Sample-Efficient Neural Architecture Search by Learning Action Space
    • 15
    • Highly Influential
    • PDF
    Progressive Neural Architecture Search
    • 870
    • PDF
    NAS-Bench-201: Extending the Scope of Reproducible Neural Architecture Search
    • 78
    • Highly Influential
    • PDF
    NAS evaluation is frustratingly hard
    • 57
    • PDF
    Efficient Neural Architecture Search via Parameter Sharing
    • 1,079
    • Highly Influential
    • PDF
    NAS-Bench-101: Towards Reproducible Neural Architecture Search
    • 147
    • Highly Influential
    • PDF
    Neural Architecture Search with Reinforcement Learning
    • 2,299
    • PDF
    Auto-DeepLab: Hierarchical Neural Architecture Search for Semantic Image Segmentation
    • 363
    • PDF