no code implementations • 27 Jun 2023 • Haitao Tang, Yu Fu, Lei Sun, Jiabin Xue, Dan Liu, Yongchao Li, Zhiqiang Ma, Minghui Wu, Jia Pan, Genshun Wan, Ming'en Zhao
In this paper, we propose an adaptive two-stage knowledge distillation method consisting of hidden layer learning and output layer learning.