Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Knowledge Distillation: A Survey

About

In recent years, deep neural networks have been successful in both industry and academia, especially for computer vision tasks. The great success of deep learning is mainly due to its scalability to encode large-scale data and to maneuver billions of model parameters. However, it is a challenge to deploy these cumbersome deep models on devices with limited resources, e.g., mobile phones and embedded devices, not only because of the high computational complexity but also the large storage requirements. To this end, a variety of model compression and acceleration techniques have been developed. As a representative type of model compression and acceleration, knowledge distillation effectively learns a small student model from a large teacher model. It has received rapid increasing attention from the community. This paper provides a comprehensive survey of knowledge distillation from the perspectives of knowledge categories, training schemes, teacher-student architecture, distillation algorithms, performance comparison and applications. Furthermore, challenges in knowledge distillation are briefly reviewed and comments on future research are discussed and forwarded.

Jianping Gou, Baosheng Yu, Stephen John Maybank, Dacheng Tao• 2020

Related benchmarks

TaskDatasetResultRank
Image ClassificationCIFAR10 (Dirichlet)
RES1
14
Image ClassificationCIFAR10 (Shards)
RES1
14
Image ClassificationCIFAR100 IID
RES100
14
Image ClassificationCIFAR100 (Dirichlet)
Robustness Score (RES)1
14
Image ClassificationCIFAR100 (Shards)
RES1
14
Image ClassificationFMNIST (IID)
RES100
14
Image ClassificationFMNIST (Dirichlet)
RES100
14
Image ClassificationFMNIST (Shards)
RES100
14
Image ClassificationEMNIST (IID)
RES100
14
Image ClassificationEMNIST Dirichlet
RES100
14
Showing 10 of 23 rows

Other info

Follow for update