End-to-end Efficient Representation Learning Via Cascading Combinatorial Optimization

Abstract

We develop hierarchically quantized efficient embedding representations for similarity-based search and show that this representation provides not only the state of the art performance on the search accuracy but also provides several orders of speed up during inference. The idea is to hierarchically quantize the representation so that the quantization granularity is greatly increased while maintaining the accuracy and keeping the computational complexity low. We also show that the problem of finding the optimal sparse compound hash code respecting the hierarchical structure can be optimized in polynomial time via minimum cost flow in an equivalent flow network. This allows us to train the method end-to-end in a mini-batch stochastic gradient descent setting. Our experiments on Cifar100 and ImageNet datasets show the state of the art search accuracy while providing several orders of magnitude search speedup respectively over exhaustive linear search over the dataset.

Stats

• citations0
• S2 citations—
• github stars0
• HF likes0
• heat score0.00
• arxiv keyjeong2019end

Related papers

• Accurate Deep Representation Quantization With Gradient Snapping Layer For Similarity Search (2016)0.00
• A Genetic Algorithm Approach For Imagerepresentation Learning Through Color Quantization (2017)0.00
• Representation Learning For Efficient And Effective Similarity Search And Recommendation (2021)0.00
• Simultaneous Compression And Quantization: A Joint Approach For Efficient Unsupervised Hashing (2018)7.50
• Interleaved Composite Quantization For High-dimensional Similarity Search (2019)0.00
• Deep Hashing Via Householder Quantization (2023)0.00
• Nearest Neighbor Search With Compact Codes: A Decoder Perspective (2021)3.58
• Learning To Compress And Search Visual Data In Large-scale Systems (2019)0.00