Explaining Knowledge Distillation by Quantifying the Knowledge

By Xu Cheng, Zhefan Rao, Yilan Chen, Quanshi Zhang. 2020

141 citations recorded by the source corpus.

This paper presents a method to interpret the success of knowledge distillation by quantifying and analyzing task-relevant and task-irrelevant visual concepts that are encoded in intermediate layers of a deep neural network (DNN). More specifically, three hypotheses are proposed as follows. 1. Knowledge distillation makes the DNN learn more visual concepts than learning from raw data. 2. Knowledge distillation ensures that the DNN is prone to learning various visual concepts simultaneously. Whereas, in the scenario of learning from raw data, the DNN learns visual concepts sequentially. 3. Knowledge distillation yields more stable optimization directions than learning from raw data. Accordingly, we design three types of mathematical metrics to evaluate feature representations of the DNN. In experiments, we diagnosed various DNNs, and above hypotheses were verified.

Subfields: Artificial Intelligence

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

Citations in the corpus

Knowledge Distillation: A Survey (2020, background): “Although the great success in practice, there are not too manyworks oneither thetheoretical or empirical understanding of knowledge distillation (Cheng et al. 2020; Phuong and Lampert 2019a; Cho and Hariharan 2019).”