A Study of Self Distillation for Mango Image Classification

Wei Chi Chen, Wei Ta Chu

Research output: Chapter in Book/Report/Conference proceedingConference contribution


We study a knowledge transfer approach called self distillation on a mango image dataset. Taking the deepest part of a convolutional neural network as the teacher, the self distillation approach transfers the relatively richer knowledge of the deepest part to shallow parts of this network, which are viewed as the students. We verify that this approach is effective in the target mango image dataset. Furthermore, we propose two more losses to improve performance considering data characteristics. In the discussion, we not only verify effectiveness of self distillation, but also point out weakness of the current approach, which unveils potential improvement for self distillation in the future.

Original languageEnglish
Title of host publicationProceedings - 2020 International Computer Symposium, ICS 2020
PublisherInstitute of Electrical and Electronics Engineers Inc.
Number of pages4
ISBN (Electronic)9781728192550
Publication statusPublished - 2020 Dec
Event2020 International Computer Symposium, ICS 2020 - Tainan, Taiwan
Duration: 2020 Dec 172020 Dec 19

Publication series

NameProceedings - 2020 International Computer Symposium, ICS 2020


Conference2020 International Computer Symposium, ICS 2020

All Science Journal Classification (ASJC) codes

  • Artificial Intelligence
  • Computer Networks and Communications
  • Computer Science Applications
  • Information Systems
  • Information Systems and Management
  • Computational Mathematics

Fingerprint Dive into the research topics of 'A Study of Self Distillation for Mango Image Classification'. Together they form a unique fingerprint.

Cite this