Accurate Deep Representation Quantization With Gradient Snapping Layer For Similarity Search

Abstract

Recent advance of large scale similarity search involves using deeply learned representations to improve the search accuracy and use vector quantization methods to increase the search speed. However, how to learn deep representations that strongly preserve similarities between data pairs and can be accurately quantized via vector quantization remains a challenging task. Existing methods simply leverage quantization loss and similarity loss, which result in unexpectedly biased back-propagating gradients and affect the search performances. To this end, we propose a novel gradient snapping layer (GSL) to directly regularize the back-propagating gradient towards a neighboring codeword, the generated gradients are un-biased for reducing similarity loss and also propel the learned representations to be accurately quantized. Joint deep representation and vector quantization learning can be easily performed by alternatively optimize the quantization codebook and the deep neural network. The pr

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyliu2016accurate

Related papers

• Spreading Vectors For Similarity Search (2018)0.00
• Interleaved Composite Quantization For High-dimensional Similarity Search (2019)0.00
• Deep Hashing Via Householder Quantization (2023)0.00
• End-to-end Efficient Representation Learning Via Cascading Combinatorial Optimization (2019)0.00
• Generalized Product Quantization Network For Semi-supervised Image Retrieval (2020)12.17
• Spatially Optimized Compact Deep Metric Learning Model For Similarity Search (2024)0.00
• Leanvec: Searching Vectors Faster By Making Them Fit (2023)0.00
• High-dimensional Similarity Search With Quantum-assisted Variational Autoencoder (2020)8.82