学习哈希自然排序
Learning to Hash Naturally Sorts
摘要：
学习散列图片列表排序问题。其测试指标，例如，平均-平均精度，对按成对代码相似性排序的排序候选列表进行计数。然而，由于排序操作的不可微性质，几乎没有人训练具有端到端排序结果的深度散列模型。训练和测试目标的这种不一致可能导致次优性能，因为训练损失通常不能反映实际的检索度量。在本文中，我们通过引入自然排序哈希（NSH）来解决这个问题。我们对样本的哈希码的汉明距离进行排序，并相应地收集它们的潜在表示进行自监督训练。由于可微排序近似的最新进展，散列头从排序器接收梯度，使得散列编码器可以与训练过程一起优化。此外，我们描述了一种新的排序噪声对比估计（SortedNCE）损失，选择性地挑选积极和消极的样本进行对比学习，这使得NSH能够在训练过程中以无监督的方式挖掘数据语义关系。我们广泛的实验表明，所提出的NSH模型显着优于现有的无监督哈希方法在三个基准数据集。
英文摘要：
Learning to hash pictures a list-wise sorting problem. Its testing metrics, e.g., mean-average precision, count on a sorted candidate list ordered by pair-wise code similarity. However, scarcely does one train a deep hashing model with the sorted results end-to-end because of the non-differentiable nature of the sorting operation. This inconsistency in the objectives of training and test may lead to sub-optimal performance since the training loss often fails to reflect the actual retrieval metric. In this paper, we tackle this problem by introducing Naturally-Sorted Hashing (NSH). We sort the Hamming distances of samples' hash codes and accordingly gather their latent representations for self-supervised training. Thanks to the recent advances in differentiable sorting approximations, the hash head receives gradients from the sorter so that the hash encoder can be optimized along with the training procedure. Additionally, we describe a novel Sorted Noise-Contrastive Estimation (SortedNCE) loss that selectively picks positive and negative samples for contrastive learning, which allows NSH to mine data semantic relations during training in an unsupervised manner. Our extensive experiments show the proposed NSH model significantly outperforms the existing unsupervised hashing methods on three benchmarked datasets.
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。
