Matches in SemOpenAlex for { <https://semopenalex.org/work/W2982425134> ?p ?o ?g. }
- W2982425134 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." @default.
- W2982425134 created "2019-11-08" @default.
- W2982425134 creator A5076540893 @default.
- W2982425134 creator A5085451600 @default.
- W2982425134 creator A5091794833 @default.
- W2982425134 date "2019-06-01" @default.
- W2982425134 modified "2023-10-11" @default.
- W2982425134 title "End-To-End Efficient Representation Learning via Cascading Combinatorial Optimization" @default.
- W2982425134 cites W1532325895 @default.
- W2982425134 cites W1921523184 @default.
- W2982425134 cites W1923967535 @default.
- W2982425134 cites W2021354639 @default.
- W2982425134 cites W2081580037 @default.
- W2982425134 cites W2113307832 @default.
- W2982425134 cites W2117539524 @default.
- W2982425134 cites W2127589108 @default.
- W2982425134 cites W2138621090 @default.
- W2982425134 cites W2143516773 @default.
- W2982425134 cites W2145287260 @default.
- W2982425134 cites W2160921898 @default.
- W2982425134 cites W2162881463 @default.
- W2982425134 cites W219040644 @default.
- W2982425134 cites W2293824885 @default.
- W2982425134 cites W2411707397 @default.
- W2982425134 cites W2549858646 @default.
- W2982425134 cites W2551835155 @default.
- W2982425134 cites W2555897561 @default.
- W2982425134 cites W2562322388 @default.
- W2982425134 cites W2594696540 @default.
- W2982425134 cites W2613997951 @default.
- W2982425134 cites W2734694826 @default.
- W2982425134 cites W2741430497 @default.
- W2982425134 cites W2798834175 @default.
- W2982425134 cites W2962830213 @default.
- W2982425134 cites W2963026686 @default.
- W2982425134 cites W2963276379 @default.
- W2982425134 cites W2963678014 @default.
- W2982425134 cites W2963911037 @default.
- W2982425134 cites W2964121744 @default.
- W2982425134 doi "https://doi.org/10.1109/cvpr.2019.01164" @default.
- W2982425134 hasPublicationYear "2019" @default.
- W2982425134 type Work @default.
- W2982425134 sameAs 2982425134 @default.
- W2982425134 citedByCount "0" @default.
- W2982425134 crossrefType "proceedings-article" @default.
- W2982425134 hasAuthorship W2982425134A5076540893 @default.
- W2982425134 hasAuthorship W2982425134A5085451600 @default.
- W2982425134 hasAuthorship W2982425134A5091794833 @default.
- W2982425134 hasConcept C111919701 @default.
- W2982425134 hasConcept C11413529 @default.
- W2982425134 hasConcept C153258448 @default.
- W2982425134 hasConcept C154945302 @default.
- W2982425134 hasConcept C173608175 @default.
- W2982425134 hasConcept C17744445 @default.
- W2982425134 hasConcept C177774035 @default.
- W2982425134 hasConcept C199539241 @default.
- W2982425134 hasConcept C206688291 @default.
- W2982425134 hasConcept C2776214188 @default.
- W2982425134 hasConcept C2776359362 @default.
- W2982425134 hasConcept C38652104 @default.
- W2982425134 hasConcept C41008148 @default.
- W2982425134 hasConcept C41608201 @default.
- W2982425134 hasConcept C50644808 @default.
- W2982425134 hasConcept C68339613 @default.
- W2982425134 hasConcept C80444323 @default.
- W2982425134 hasConcept C94625758 @default.
- W2982425134 hasConcept C99138194 @default.
- W2982425134 hasConceptScore W2982425134C111919701 @default.
- W2982425134 hasConceptScore W2982425134C11413529 @default.
- W2982425134 hasConceptScore W2982425134C153258448 @default.
- W2982425134 hasConceptScore W2982425134C154945302 @default.
- W2982425134 hasConceptScore W2982425134C173608175 @default.
- W2982425134 hasConceptScore W2982425134C17744445 @default.
- W2982425134 hasConceptScore W2982425134C177774035 @default.
- W2982425134 hasConceptScore W2982425134C199539241 @default.
- W2982425134 hasConceptScore W2982425134C206688291 @default.
- W2982425134 hasConceptScore W2982425134C2776214188 @default.
- W2982425134 hasConceptScore W2982425134C2776359362 @default.
- W2982425134 hasConceptScore W2982425134C38652104 @default.
- W2982425134 hasConceptScore W2982425134C41008148 @default.
- W2982425134 hasConceptScore W2982425134C41608201 @default.
- W2982425134 hasConceptScore W2982425134C50644808 @default.
- W2982425134 hasConceptScore W2982425134C68339613 @default.
- W2982425134 hasConceptScore W2982425134C80444323 @default.
- W2982425134 hasConceptScore W2982425134C94625758 @default.
- W2982425134 hasConceptScore W2982425134C99138194 @default.
- W2982425134 hasLocation W29824251341 @default.
- W2982425134 hasOpenAccess W2982425134 @default.
- W2982425134 hasPrimaryLocation W29824251341 @default.
- W2982425134 hasRelatedWork W2101618614 @default.
- W2982425134 hasRelatedWork W2259373126 @default.
- W2982425134 hasRelatedWork W2410805313 @default.
- W2982425134 hasRelatedWork W2799167061 @default.
- W2982425134 hasRelatedWork W2803553068 @default.
- W2982425134 hasRelatedWork W2894162850 @default.
- W2982425134 hasRelatedWork W2898992238 @default.
- W2982425134 hasRelatedWork W2911440615 @default.
- W2982425134 hasRelatedWork W2917392503 @default.
- W2982425134 hasRelatedWork W2951427637 @default.
- W2982425134 hasRelatedWork W2986514577 @default.