Abstract | ||
---|---|---|
Lifelong learning with deep neural networks is well-known to suffer from catastrophic forgetting: the performance on previous tasks drastically degrades when learning a new task. To alleviate this effect, we propose to leverage a large stream of unlabeled data easily obtainable in the wild. In particular, we design a novel class-incremental learning scheme with (a) a new distillation loss, termed global distillation, (b) a learning strategy to avoid overfitting to the most recent task, and (c) a confidence-based sampling method to effectively leverage unlabeled external data. Our experimental results on various datasets, including CIFAR and ImageNet, demonstrate the superiority of the proposed methods over prior methods, particularly when a stream of unlabeled data is accessible: our method shows up to 15.8% higher accuracy and 46.5% less forgetting compared to the state-of-the-art method. The code is available at https://github.com/kibok90/iccv2019-inc. |
Year | DOI | Venue |
---|---|---|
2019 | 10.1109/ICCV.2019.00040 | 2019 IEEE/CVF International Conference on Computer Vision (ICCV) |
Keywords | Field | DocType |
distillation loss,global distillation,learning strategy,catastrophic forgetting,lifelong learning,deep neural networks,class-incremental learning scheme,unlabeled external data,confidence-based sampling method,CIFAR dataset,ImageNet dataset | Forgetting,Pattern recognition,Computer science,Speech recognition,Artificial intelligence | Conference |
Volume | Issue | ISSN |
2019 | 1 | 1550-5499 |
ISBN | Citations | PageRank |
978-1-7281-4804-5 | 7 | 0.49 |
References | Authors | |
7 | 4 |
Name | Order | Citations | PageRank |
---|---|---|---|
Kibok Lee | 1 | 68 | 5.16 |
Kimin Lee | 2 | 51 | 11.57 |
Jinwoo Shin | 3 | 513 | 56.35 |
Honglak Lee | 4 | 6247 | 398.39 |