Hashing as Tie-Aware Learning to Rank

Kun He, Fatih Cakir, Sarah Adel Bargal, Stan Sclaroff; Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR), 2018, pp. 4023-4032

Abstract


Hashing, or learning binary embeddings of data, is frequently used in nearest neighbor retrieval. In this paper, we develop learning to rank formulations for hashing, aimed at directly optimizing ranking-based evaluation metrics such as Average Precision (AP) and Normalized Discounted Cumulative Gain (NDCG). We first observe that the integer-valued Hamming distance often leads to tied rankings, and propose to use tie-aware versions of AP and NDCG to evaluate hashing for retrieval. Then, to optimize tie-aware ranking metrics, we derive their continuous relaxations, and perform gradient-based optimization with deep neural networks. Our results establish the new state-of-the-art for image retrieval by Hamming ranking in common benchmarks.

Related Material


[pdf] [supp] [arXiv]
[bibtex]
@InProceedings{He_2018_CVPR,
author = {He, Kun and Cakir, Fatih and Bargal, Sarah Adel and Sclaroff, Stan},
title = {Hashing as Tie-Aware Learning to Rank},
booktitle = {Proceedings of the IEEE Conference on Computer Vision and Pattern Recognition (CVPR)},
month = {June},
year = {2018}
}