Optimization Of Rank Losses For Image Retrieval

Abstract

In image retrieval, standard evaluation metrics rely on score ranking, \eg average precision (AP), recall at k (R@k), normalized discounted cumulative gain (NDCG). In this work we introduce a general framework for robust and decomposable rank losses optimization. It addresses two major challenges for end-to-end training of deep neural networks with rank losses: non-differentiability and non-decomposability. Firstly we propose a general surrogate for ranking operator, SupRank, that is amenable to stochastic gradient descent. It provides an upperbound for rank losses and ensures robust training. Secondly, we use a simple yet effective loss function to reduce the decomposability gap between the averaged batch approximation of ranking losses and their values on the whole training set. We apply our framework to two standard metrics for image retrieval: AP and R@k. Additionally we apply our framework to hierarchical image retrieval. We introduce an extension of AP, the hierarchical average p

Stats

• citations2
• S2 citations—
• github stars845
• HF likes0
• heat score9.43
• arxiv keyramzi2023optimization

Related papers

• Learning With Average Precision: Training Image Retrieval With A Listwise Loss (2019)18.86
• Robust And Decomposable Average Precision For Image Retrieval (2021)0.00
• Optimizing Rank-based Metrics With Blackbox Differentiation (2019)17.35
• Smooth-ap: Smoothing The Path Towards Large-scale Image Retrieval (2020)15.06
• Sodeep: A Sorting Deep Net To Learn Ranking Loss Surrogates (2019)12.33
• Hierarchical Average Precision Training For Pertinent Image Retrieval (2022)9.10
• Hashing As Tie-aware Learning To Rank (2017)13.11
• An Alternative Cross Entropy Loss For Learning-to-rank (2019)11.58