DeeSIL：深度-浅层增量学习
DeeSIL: Deep-Shallow Incremental Learning
摘要：
增量学习（IL）是一个有趣的AI问题，当算法被假设为在预算上工作时。当使用深度学习方法对IL进行建模时尤其如此，其中由于有限的记忆而出现两个复杂的挑战，这引起了与为了并入新类所需的再训练相关的灾难性遗忘和延迟。在这里，我们介绍DeeSIL，一种已知的迁移学习方案的适应，其组合了用作特征提取器的固定深度表示和学习独立的浅分类器以增加识别能力。该方案解决了上述两个挑战，因为它在有限的内存预算下工作得很好，并且每个新概念都可以在一分钟内添加。此外，由于模型递增时不需要深度重新训练，DeeSIL可以集成更大量的初始数据，提供更多可转移的特征。在ImageNet LSVRC 2012上对三种最先进的算法进行了性能评估。结果表明，在大规模下，使用相同和更多的初始数据时，DeeSIL性能分别比最佳基线高出23和33个点。
英文摘要：
Incremental Learning (IL) is an interesting AI problem when the algorithm is assumed to work on a budget. This is especially true when IL is modeled using a deep learning approach, where two com- plex challenges arise due to limited memory, which induces catastrophic forgetting and delays related to the retraining needed in order to incorpo- rate new classes. Here we introduce DeeSIL, an adaptation of a known transfer learning scheme that combines a fixed deep representation used as feature extractor and learning independent shallow classifiers to in- crease recognition capacity. This scheme tackles the two aforementioned challenges since it works well with a limited memory budget and each new concept can be added within a minute. Moreover, since no deep re- training is needed when the model is incremented, DeeSIL can integrate larger amounts of initial data that provide more transferable features. Performance is evaluated on ImageNet LSVRC 2012 against three state of the art algorithms. Results show that, at scale, DeeSIL performance is 23 and 33 points higher than the best baseline when using the same and more initial data respectively.
机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。
