Knowledge Representing: Efficient, Sparse Representation of Prior Knowledge for Knowledge Distillation

By Junjie Liu, Dongchao Wen, Hongxing Gao, Wei Tao, Tse-Wei Chen, Kinya Osa, Masami Kato. 2019

27 citations recorded by the source corpus.

Despite the recent works on knowledge distillation (KD) have achieved a further improvement through elaborately modeling the decision boundary as the posterior knowledge, their performance is still dependent on the hypothesis that the target network has a powerful capacity (representation ability). In this paper, we propose a knowledge representing (KR) framework mainly focusing on modeling the parameters distribution as prior knowledge. Firstly, we suggest a knowledge aggregation scheme in order to answer how to represent the prior knowledge from teacher network. Through aggregating the parameters distribution from teacher network into more abstract level, the scheme is able to alleviate the phenomenon of residual accumulation in the deeper layers. Secondly, as the critical issue of what the most important prior knowledge is for better distilling, we design a sparse recoding penalty for constraining the student network to learn with the penalized gradients. With the proposed penalty, the student network can effectively avoid the over-regularization during knowledge distilling and converge faster. The quantitative experiments exhibit that the proposed framework achieves the state-ofthe-arts performance, even though the target network does not have the expected capacity. Moreover, the framework is flexible enough for combining with other KD methods based on the posterior knowledge.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

1. Dacheng Tao
2. Can Xu
3. Jianping Gou
4. S. Maybank
5. Baosheng Yu
6. Daxin Jiang
7. Xiuyi Chen
8. Zhe Gan

In Computer Vision and Pattern Recognition

1. Dacheng Tao
2. Can Xu
3. Daxin Jiang
4. S. Maybank
5. Jianping Gou
6. Xiuyi Chen
7. Baosheng Yu
8. Luc Van Gool

Citations in the corpus

• Knowledge Distillation: A Survey (2020, methodology): “Furthermore, the parameters of the teacher model (or the connections between layers) also contain another knowledge (Liu et al. 2019c).”
• Learning to Ground Visual Objects for Visual Dialog (2021, methodology): “Knowledge Distillation Knowledge distillation (Hinton et al., 2015; Gou et al., 2020; Mirzadeh et al., 2020; Romero et al., 2014; Liu et al., 2019; Jiao et al., 2019) has been widely used in different fields of artificial intelligence, including visual recognition, speech, recognition, natural langu”