Modeling Teacher-Student Techniques in Deep Neural Networks for Knowledge Distillation
This work provides a foundational framework for researchers in machine learning to understand and compare KD techniques, though it is incremental as it synthesizes existing methods rather than introducing new ones.
The paper tackles the lack of a generalized model for knowledge distillation (KD) techniques by investigating and analyzing various studies to build a general model that summarizes all KD methods, enabling better investigation and exploration of different approaches.
Knowledge distillation (KD) is a new method for transferring knowledge of a structure under training to another one. The typical application of KD is in the form of learning a small model (named as a student) by soft labels produced by a complex model (named as a teacher). Due to the novel idea introduced in KD, recently, its notion is used in different methods such as compression and processes that are going to enhance the model accuracy. Although different techniques are proposed in the area of KD, there is a lack of a model to generalize KD techniques. In this paper, various studies in the scope of KD are investigated and analyzed to build a general model for KD. All the methods and techniques in KD can be summarized through the proposed model. By utilizing the proposed model, different methods in KD are better investigated and explored. The advantages and disadvantages of different approaches in KD can be better understood and develop a new strategy for KD can be possible. Using the proposed model, different KD methods are represented in an abstract view.