ACIL: analytic class-incremental learning with absolute memorization and privacy protection
Class-incremental learning (CIL) learns a classification model with training data of different classes arising progressively. Existing CIL either suffers from serious accuracy loss due to catastrophic forgetting, or invades data privacy by revisiting used exemplars. Inspired by linear learning fo...
Saved in:
| Main Authors: | , , , , |
|---|---|
| 其他作者: | |
| 格式: | Conference or Workshop Item |
| 語言: | English |
| 出版: |
2024
|
| 主題: | |
| 在線閱讀: | https://hdl.handle.net/10356/174481 https://proceedings.neurips.cc/paper_files/paper/2022 |
| 標簽: |
添加標簽
沒有標簽, 成為第一個標記此記錄!
|
| 總結: | Class-incremental learning (CIL) learns a classification model with training data
of different classes arising progressively. Existing CIL either suffers from serious
accuracy loss due to catastrophic forgetting, or invades data privacy by revisiting
used exemplars. Inspired by linear learning formulations, we propose an analytic
class-incremental learning (ACIL) with absolute memorization of past knowledge
while avoiding breaching of data privacy (i.e., without storing historical data).
The absolute memorization is demonstrated in the sense that class-incremental
learning using ACIL given present data would give identical results to that from
its joint-learning counterpart which consumes both present and historical samples.
This equality is theoretically validated. Data privacy is ensured since no historical
data are involved during the learning process. Empirical validations demonstrate
ACIL’s competitive accuracy performance with near-identical results for various
incremental task settings (e.g., 5-50 phases). This also allows ACIL to outperform
the state-of-the-art methods for large-phase scenarios (e.g., 25 and 50 phases). |
|---|