arXivDaily arXiv每日学术速递 周一至周五更新
arXiv 1810.01008cs.LGcs.CVstat.ML
收藏

基于Hamming距离目标的Hash码学习

Learning Hash Codes via Hamming Distance Targets

Hive

Martin Loncaric, Bowei Liu, Ryan Weber

摘要:

我们提出了一个强大的新的损失函数和训练计划学习二进制哈希码与任何可微模型和相似性函数。我们的损失函数通过使用对数似然损失对两个输入落在汉明距离目标内的概率进行精确近似来改进先前的方法。我们的新训练方案通过更好地采样输入和评估每个minibatch中所有输入对之间的损失项来获得真实梯度的良好估计。为了充分利用产生的哈希,我们使用多索引。我们证明,这些技术提供了很大的改进相似性搜索任务。我们报告了迄今为止ImageNet和SIFT 1 M在竞争信息检索任务上的最佳结果,将MAP从73%提高到84%,并将查询成本降低了2-8倍。

英文摘要:

We present a powerful new loss function and training scheme for learning binary hash codes with any differentiable model and similarity function. Our loss function improves over prior methods by using log likelihood loss on top of an accurate approximation for the probability that two inputs fall within a Hamming distance target. Our novel training scheme obtains a good estimate of the true gradient by better sampling inputs and evaluating loss terms between all pairs of inputs in each minibatch. To fully leverage the resulting hashes, we use multi-indexing. We demonstrate that these techniques provide large improvements to a similarity search tasks. We report the best results to date on competitive information retrieval tasks for ImageNet and SIFT 1M, improving MAP from 73% to 84% and reducing query cost by a factor of 2-8, respectively.

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。

↑