Publications
금융권 유일의 연구 조직으로 다양한 신기술 영역에서 하나금융그룹의 위상을 높이고
세계적 권위의 학회에서 대외 성과를 달성하고 있습니다.
Papers
Differentiable Neural Architecture Transformation For Reproducible Architecture Improvement
- 발행일
- 2020.06.15
- 발행기관
- arXiv(2020)
- 저자
- Do-Guk Kim, Heung-Chang Lee
- Link
- https://arxiv.org/abs/2006.08231
초록
Recently, Neural Architecture Search (NAS) methods are introduced and show impressive performance on many benchmarks. Among those NAS studies, Neural Architecture Transformer (NAT) aims to improve the given neural architecture to have better performance while maintaining computational costs. However, NAT has limitations about a lack of reproducibility. In this paper, we propose differentiable neural architecture transformation that is reproducible and efficient. The proposed method shows stale performance on various architectures. Extensive reproducibility experiments on two datasets, i.e., CIFAR-10 and Tiny Imagenet, present that the proposed method definitely outperforms NAT and be applicable to other models and datasets
PDF View