A Simple Class Decision Balancing for Incremental Learning

By Hongjoon Ahn, Taesup Moon. 2020

7 citations recorded by the source corpus.

How researchers see this paper

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

1. Eden Belouadah
2. Adrian Daniel Popescu
3. Ioannis Kanellos
4. Sudhanshu Mittal
5. Silvio Galesso
6. Thomas Brox
7. Jongyoon Kim
8. Da-Wei Zhou

Citations in the corpus

• Split-and-Bridge: Adaptable Class Incremental Learning within a Single Neural Network (2021, methodology): “The original version of iCaRL is slightly different from STD, but this STD version is often reported to perform even better than the original one by the recent works such as (Ahn and Moon 2020).”
• Essentials for Class Incremental Learning (2021, methodology): “A similar analysis has been shown by [1] for a fine-tuning setup.”
• Initial Classifier Weights Replay for Memoryless Class Incremental Learning (2020, background): “One influential finding from [22] was that vanilla fine tuning (FT ) is unfit for IL but this finding was recently challenged for IL with bounded memory of the past [1, 5].”