Autoencoder-Based Incremental Class Learning without Retraining on Old Data

By Euntae Choi, Kyungmi Lee, Kiyoung Choi. 2019

13 citations recorded by the source corpus.

Incremental class learning, a scenario in continual learning context where classes and their training data are sequentially and disjointedly observed, challenges a problem widely known as catastrophic forgetting. In this work, we propose a novel incremental class learning method that can significantly reduce memory overhead compared to previous approaches. Apart from conventional classification scheme using softmax, our model bases on an autoencoder to extract prototypes for given inputs so that no change in its output unit is required. It stores only the mean of prototypes per class to perform metric-based classification, unlike rehearsal approaches which rely on large memory or generative model. To mitigate catastrophic forgetting, regularization methods are applied on our model when a new task is encountered. We evaluate our method by experimenting on CIFAR-100 and CUB-200-2011 and show that its performance is comparable to the state-of-the-art method with much lower additional memory cost.

Subfields: Artificial Intelligence, Computer Vision and Pattern Recognition, Radiology, Nuclear Medicine and Imaging

How researchers see this paper

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

In Artificial Intelligence

1. Xinchao Wang
2. Mingli Song
3. Jingwen Ye
4. Jie Song
5. Richard E. Turner
6. Xingyi Yang
7. Tameem Adel
8. Han Zhao

In Computer Vision and Pattern Recognition

1. Xinchao Wang
2. Mingli Song
3. Jingwen Ye
4. Jie Song
5. Xingyi Yang
6. Tameem Adel
7. Richard E. Turner
8. Songhua Liu

In Radiology, Nuclear Medicine and Imaging

1. Xinchao Wang
2. Mingli Song
3. Richard E. Turner
4. Tameem Adel
5. Han Zhao

Citations in the corpus

• Learning with Recoverable Forgetting (2022, positive): “Apart from those works that still need part of the old data, many researchers are devoted to developing the methods without storing the old data by synthesizing old data [9,41,46] or even without referring to any old data [44,42,32].”
• Continual Learning with Adaptive Weights (CLAW) (2019, background): “A major limitation of the memory-based approach is that data from previous tasks may not be available in all real-world problems (Shin et al., 2017; Choi et al., 2019).”