Variational Bayesian Multiple Instance Learning with Gaussian Processes
Variational Bayesian Multiple Instance Learning with Gaussian Processes
复制标题
DOI：
10.1109/cvpr.2017.93
复制
发表时间：
2017-07
期刊：
影响因子：
--
通讯作者：
Manuel Haussmann;F. Hamprecht;M. Kandemir
中科院分区：
文献类型：
--
作者：
Manuel Haussmann;F. Hamprecht;M. Kandemir
高斯过程（GP）是有效的贝叶斯预测器。我们在这里首次表明，GP分类器的实例标签可以在多实例学习（MIL）设置使用变分贝叶斯推断。我们通过一个新的建设袋的可能性，假设一个大的值，如果实例预测遵守MIL的约束和一个小的值，否则，我们实现了这一点。这种构造使我们能够解析地导出变分参数的更新规则，从而确保可扩展学习和快速收敛。我们观察到这个模型，以提高最先进的实例标签预测袋级监督在20个新闻组基准，以及在巴雷特癌症肿瘤定位从组织病理学组织微阵列图像。此外，我们引入了一个新的弱监督对象检测管道，自然地补充了我们的模型，这提高了PASCAL VOC 2007和2012数据集的最新技术水平。最后但并非最不重要的是，我们的模型的性能可以通过使用混合监督来进一步提高：弱（包）和强（实例）标签的组合。
Gaussian Processes (GPs) are effective Bayesian predictors. We here show for the first time that instance labels of a GP classifier can be inferred in the multiple instance learning (MIL) setting using variational Bayes. We achieve this via a new construction of the bag likelihood that assumes a large value if the instance predictions obey the MIL constraints and a small value otherwise. This construction lets us derive the update rules for the variational parameters analytically, assuring both scalable learning and fast convergence. We observe this model to improve the state of the art in instance label prediction from bag-level supervision in the 20 Newsgroups benchmark, as well as in Barretts cancer tumor localization from histopathology tissue microarray images. Furthermore, we introduce a novel pipeline for weakly supervised object detection naturally complemented with our model, which improves the state of the art on the PASCAL VOC 2007 and 2012 data sets. Last but not least, the performance of our model can be further boosted up using mixed supervision: a combination of weak (bag) and strong (instance) labels.