Summary of Joint Input and Output Coordination For Class-incremental Learning, by Shuai Wang et al.
Joint Input and Output Coordination for Class-Incremental Learning
by Shuai Wang, Yibing Zhan, Yong Luo, Han Hu, Wei Yu, Yonggang Wen, Dacheng Tao
First submitted to arxiv on: 9 Sep 2024
Categories
- Main: Machine Learning (cs.LG)
- Secondary: Artificial Intelligence (cs.AI)
GrooveSquid.com Paper Summaries
GrooveSquid.com’s goal is to make artificial intelligence research accessible by summarizing AI papers in simpler terms. Each summary below covers the same AI paper, written at different levels of difficulty. The medium difficulty and low difficulty versions are original summaries written by GrooveSquid.com, while the high difficulty version is the paper’s original abstract. Feel free to learn from the version that suits you best!
| Summary difficulty | Written by | Summary |
|---|---|---|
| High | Paper authors | High Difficulty Summary Read the original abstract here |
| Medium | GrooveSquid.com (original content) | Medium Difficulty Summary The paper proposes a joint input and output coordination (JIOC) mechanism to address the challenges of catastrophic forgetting in incremental learning. This mechanism assigns weights to data categories based on output scores and uses knowledge distillation to reduce interference between old and new tasks. The approach is generalizable and can be incorporated into various memory-based incremental learning methods. |
| Low | GrooveSquid.com (original content) | Low Difficulty Summary In this paper, researchers try to solve a big problem called “catastrophic forgetting” in machine learning. When we learn new things, we often forget the old ones. The authors want to find a way to remember both new and old information at the same time. They propose an idea that helps us decide which old data is most important to keep, based on how well it predicts what we’re trying to do. This idea also helps reduce the negative effects of learning something new. |
Keywords
» Artificial intelligence » Knowledge distillation » Machine learning
