Stochastic Precision Ensemble: Self-Knowledge Distillation for Quantized Deep Neural Networks
- Title
- Stochastic Precision Ensemble: Self-Knowledge Distillation for Quantized Deep Neural Networks
- Author
- 최정욱
- Issue Date
- 2021-02
- Publisher
- Association for the Advancement of Artificial Intelligence
- Citation
- 35th AAAI Conference on Artificial Intelligence, v. 35, page. 6794-6802
- Abstract
- The quantization of deep neural networks (QDNNs) has been actively studied for deployment in edge devices. Recent studies employ the knowledge distillation (KD) method to improve the performance of quantized networks. In this study, we propose stochastic precision ensemble training for
QDNNs (SPEQ). SPEQ is a knowledge distillation training scheme; however, the teacher is formed by sharing the model parameters of the student network. We obtain the soft labels of the teacher by randomly changing the bit precision of the activation stochastically at each layer of the forward-pass computation. The student model is trained with these soft labels to reduce the activation quantization noise. The cosine similarity loss is employed, instead of the KL-divergence, for KD training. As the teacher model changes continuously by random bit-precision assignment, it exploits the effect of stochastic ensemble KD. SPEQ outperforms the existing quantization training methods in various tasks, such as image classification, question-answering, and transfer learning without the need for cumbersome teacher networks.
- URI
- https://ojs.aaai.org/index.php/AAAI/article/view/16839https://repository.hanyang.ac.kr/handle/20.500.11754/176246
- ISSN
- 2159-5399; 2374-3468
- DOI
- 10.1609/aaai.v35i8.16839
- Appears in Collections:
- COLLEGE OF ENGINEERING[S](공과대학) > ELECTRONIC ENGINEERING(융합전자공학부) > Articles
- Files in This Item:
There are no files associated with this item.
- Export
- RIS (EndNote)
- XLS (Excel)
- XML