Matching-oriented Product Quantization For Ad-hoc Retrieval

Abstract

Product quantization (PQ) is a widely used technique for ad-hoc retrieval. Recent studies propose supervised PQ, where the embedding and quantization models can be jointly trained with supervised learning. However, there is a lack of appropriate formulation of the joint training objective; thus, the improvements over previous non-supervised baselines are limited in reality. In this work, we propose the Matching-oriented Product Quantization (MoPQ), where a novel objective Multinoulli Contrastive Loss (MCL) is formulated. With the minimization of MCL, we are able to maximize the matching probability of query and ground-truth key, which contributes to the optimal retrieval accuracy. Given that the exact computation of MCL is intractable due to the demand of vast contrastive samples, we further propose the Differentiable Cross-device Sampling (DCS), which significantly augments the contrastive samples for precise approximation of MCL. We conduct extensive experimental studies on four real

Stats

• citations0
• S2 citations—
• github stars13
• HF likes0
• heat score2.29
• arxiv keyxiao2021matching

Related papers

• Jointly Optimizing Query Encoder And Product Quantization To Improve Retrieval Performance (2021)12.74
• End-to-end Supervised Product Quantization For Image Search And Retrieval (2017)13.05
• Beyond Product Quantization: Deep Progressive Quantization For Image Retrieval (2019)12.95
• Self-supervised Product Quantization For Deep Unsupervised Image Retrieval (2021)13.44
• Online Product Quantization (2017)10.61
• Generalized Product Quantization Network For Semi-supervised Image Retrieval (2020)12.17
• Adversarial Attack On Deep Product Quantization Network For Image Retrieval (2020)11.29
• Hihpq: Hierarchical Hyperbolic Product Quantization For Unsupervised Image Retrieval (2024)6.77