具有模糊任务边界和噪声标签的数据流采样
Data Stream Sampling with Fuzzy Task Boundaries and Noisy Labels
AI总结：
针对模糊任务边界与噪声标签的数据流持续学习问题，提出 NTD 采样去偏方法，在四个数据集上以更少 GPU 内存和两倍以上加速保持或提升准确率。
摘要：
在持续学习领域，数据流中噪声标签的存在是模型可靠性和公平性的一个显著障碍。我们关注相关文献中概述的数据流场景，其特征是模糊的任务边界和噪声标签。为应对这一挑战，我们提出了一种新颖且直观的采样方法，称为 Noisy Test Debiasing（NTD，噪声测试去偏），以缓解不断演化的数据流中的噪声标签，并建立公平且鲁棒的持续学习算法。NTD 实现简单，使其能够适用于各种场景。我们的实验在四个数据集上进行基准测试，包括两个合成噪声数据集（CIFAR10 和 CIFAR100）以及真实世界噪声数据集（mini-WebVision 和 Food-101N）。结果验证了 NTD 在数据流含噪声标签场景下用于在线持续学习的有效性。与此前领先的方法相比，NTD 实现了两倍以上的训练加速，同时保持或超越了准确率水平。此外，与此前领先方法相比，NTD 使用的 GPU 内存资源不到五分之一。
英文摘要：
In the realm of continual learning, the presence of noisy labels within data streams represents a notable obstacle to model reliability and fairness. We focus on the data stream scenario outlined in pertinent literature, characterized by fuzzy task boundaries and noisy labels. To address this challenge, we introduce a novel and intuitive sampling method called Noisy Test Debiasing (NTD) to mitigate noisy labels in evolving data streams and establish a fair and robust continual learning algorithm. NTD is straightforward to implement, making it feasible across various scenarios. Our experiments benchmark four datasets, including two synthetic noise datasets (CIFAR10 and CIFAR100) and real-world noise datasets (mini-WebVision and Food-101N). The results validate the efficacy of NTD for online continual learning in scenarios with noisy labels in data streams. Compared to the previous leading approach, NTD achieves a training speedup enhancement over two times while maintaining or surpassing accuracy levels. Moreover, NTD utilizes less than one-fifth of the GPU memory resources compared to previous leading methods.
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。
