Mitigating Forgetting in Online Continual Learning with Neuron Calibration

By Haiyan Yin, Peng Yang. 2021

38 citations recorded by the source corpus.

Subfields: Meta & Continual Learning

How researchers see this paper

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

In Meta & Continual Learning

1. Jorge Armando Mendez Mendez
2. Eric Eaton
3. Steven C. H. Hoi
4. B. Liu
5. Chenghao Liu
6. Quang Pham
7. Bing Liu
8. Dongyan Zhao

Citations in the corpus

• Incremental Task Learning with Incremental Rank Updates (2022, background): “It has been a popular continual learning setup even in the very recent literature [7,14,11,36,31,40].”
• Online Continual Learning through Mutual Information Maximization (2022, background): “NCCL (Yin et al., 2021) calibrates the network to mitigate CF.”
• Continual Learning, Fast and Slow (2021, background): “Particularly, they introduce a separate controller to calibrate the backbone network to balance between alleviating forgetting and facilitating knowledge transfer [43], [44].”
• How to Reuse and Compose Knowledge for a Lifetime of Tasks: A Survey on Continual Learning and Functional Composition (2022, background): “, 2017) with functional regularization to avoid forgetting in an approach based on weight and feature calibration (Yin et al., 2021).”