Current deep learning paradigms largely benefit from the tremendous amount of annotated data. However, the quality of the annotations often varies among labelers. Multi-observer studies have been conducted to study these annotation variances (by labeling the same data for multiple times) and its effects on critical applications like medical image analysis. This process indeed adds an extra burden to the already tedious annotation work that usually requires professional training and expertise in the specific domains. On the other hand, automated annotation methods based on NLP algorithms have recently shown promise as a reasonable alternative, relying on the existing diagnostic reports of those images that are widely available in the clinical system. Compared to human labelers, different algorithms provide labels with varying qualities that are even noisier. In this paper, we show how noisy annotations (e.g., from different algorithm-based labelers) can be utilized together and mutually benefit the learning of classification tasks. Specifically, the concept of attention-on-label is introduced to sample better label sets on-the-fly as the training data. A meta-training based label-sampling module is designed to attend the labels that benefit the model learning the most through additional back-propagation processes. We apply the attention-on-label scheme on the classification task of a synthetic noisy CIFAR-10 dataset to prove the concept, and then demonstrate superior results (3-5% increase on average in multiple disease classification AUCs) on the chest x-ray images from a hospital-scale dataset (MIMIC-CXR) and hand-labeled dataset (OpenI) in comparison to regular training paradigms.
On-the-Fly Guidance Training for Medical Image Registration
University of Leeds(利兹大学);University of California, Irvine(加州大学欧文分校);Tongji University(同济大学);Huazhong University of Science and Technology(华中科技大学);Fudan Univeristy(复旦大学);University of California, San Diego(加州大学圣迭戈分校)
Robust and On-the-fly Dataset Denoising for Image Classification
Stanford University(斯坦福大学);Facebook AI Research(Facebook人工智能研究院);Google Brain(谷歌大脑)
过度参数化的神经网络中的泛化在错误标记的例子中会严重损害泛化能力。然而,错误标记的例子是很难避免的,在非常大的数据集收集弱监督。我们通过反事实地推理具有均匀随机标签的示例的损失分布来解决这个问题,如果它们是用真实示例训练的,并使用这些信息从训练集中去除噪声示例。首先,我们观察到,在大学习率下使用随机梯度下降训练时,具有均匀随机标签的示例具有更高的损失。然后,我们建议只使用网络参数对反事实示例的损失分布进行建模,这能够对此类示例进行建模,并取得了显着的成功。最后,我们建议删除损失超过建模损失分布的某个分位数的示例。这导致了On-the-fly Data Denoising(ODD),这是一种简单而有效的算法,对错误标记的示例具有鲁棒性,同时与标准训练相比,几乎没有计算开销。ODD能够在广泛的数据集上实现最先进的结果,包括真实世界的数据集,如WebVision和Clothing 1 M。
Large Loss Matters in Weakly Supervised Multi-Label Classification
Seoul National University(首尔国立大学);University of Tübingen(图宾根大学);Max Planck Institute for Intelligent Systems(马克斯·普朗克智能系统研究所);Max Planck Institute for Informatics(马克斯·普朗克信息学研究所);HodooAI Lab(HodooAI实验室)
Learning Disentangled Label Representations for Multi-label Classification
University of Chinese Academy of Sciences(中国科学院大学);Institute of Automation, Chinese Academy of Sciences(中国科学院自动化研究所);CAS Center for Excellence in Brain Science and Intelligence Technology(中国科学院脑科学与智能技术卓越创新中心)
VRFP: On-the-fly Video Retrieval using Web Images and Fast Fisher Vector Products
University of Maryland, College Park(马里兰大学帕克分校)
VRFP是一个基于短文本输入查询的实时视频检索框架,它在查询已知后从Web上获取弱标记的训练图像。检索到的表示查询的Web图像和每个数据库视频被视为无序的图像集合,并且每个集合使用基于CNN特征构建的单个Fisher向量来表示。我们的实验表明,Fisher向量对网络图像中存在的噪声具有鲁棒性,并且在准确性方面优于其他标准表示。虽然可以为新查询有效地构建Fisher向量,但由于其高维性,与测试集的匹配很慢。为了进行实时匹配,我们提出了一种无损算法,加速高维Fisher向量之间的内积计算。我们证明了所需的乘法的预期数量减少二次与稀疏的Fisher向量。我们不仅能够实时构建和应用查询模型,而且在一个简单的重新排序方案的帮助下,我们在TRECVID MED 13(3.5%),MED 14(1.3%)和CCV数据集(5.2%)上的表现也明显优于最先进的自动检索方法。我们还提供了一个直接的比较标准的数据集之间的两个不同的范例自动视频检索-零样本学习和飞行检索。
Bridging the Gap between Model Explanations in Partially Annotated Multi-label Classification
Seoul National University(首尔国立大学);University of Tübingen(图宾根大学);HodooAI Lab(HodooAI实验室);Inria(法国国家信息与自动化研究所);Ecole normale supérieure(巴黎高等师范学院);CNRS(法国国家科学研究中心);PSL Research University(巴黎文理研究大学);MPI for Intelligent Systems(马克斯·普朗克智能系统研究所)