arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 1512.04785cs.CVcs.MM
收藏

基于噪声网络图像的深度表示学习

On Deep Representation Learning from Noisy Web Images

CEA LIST(法国原子能与替代能源委员会信息实验室)

Phong D. Vo, Alexandru Ginsca, Hervé Le Borgne, Adrian Popescu

摘要:

不断增长的Web图像内容可能是扩展深度神经网络的下一个重要数据源,最近在ImageNet分类挑战和相关任务中取得了巨大成功。然而,由于其监督机制,这种前景尚未在卷积网络(convnet)上得到验证-这是表现最好的深度模型之一。虽然无监督替代方案在将学习模型推广到新领域方面不如convnet好,但我们使用convnet来利用半监督表示学习。我们的方法是使用大量的未标记和嘈杂的Web图像训练convnets作为一般的特征检测器,尽管来自数据的挑战,如高水平的错误标记的数据,离群值,和数据偏差。在几个数据规模,不同的网络架构和数据重新排序技术进行了广泛的实验。在各种主题的九个公共数据集上评估学习的表示。我们在314万张Web图像上训练的convnets获得的最佳结果优于在ILSVRC 2012的120万张干净图像上训练的AlexNet,并且正在缩小与VGG-16的差距。这些突出的结果表明了在实践中使用深度学习的预算解决方案,并激励了半监督表示学习的更多研究。

英文摘要:

The keep-growing content of Web images may be the next important data source to scale up deep neural networks, which recently obtained a great success in the ImageNet classification challenge and related tasks. This prospect, however, has not been validated on convolutional networks (convnet) -- one of best performing deep models -- because of their supervised regime. While unsupervised alternatives are not so good as convnet in generalizing the learned model to new domains, we use convnet to leverage semi-supervised representation learning. Our approach is to use massive amounts of unlabeled and noisy Web images to train convnets as general feature detectors despite challenges coming from data such as high level of mislabeled data, outliers, and data biases. Extensive experiments are conducted at several data scales, different network architectures, and data reranking techniques. The learned representations are evaluated on nine public datasets of various topics. The best results obtained by our convnets, trained on 3.14 million Web images, outperform AlexNet trained on 1.2 million clean images of ILSVRC 2012 and is closing the gap with VGG-16. These prominent results suggest a budget solution to use deep learning in practice and motivate more research in semi-supervised representation learning.

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。

↑