FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W3171091521> ?p ?o ?g. }

• W3171091521 abstract "Learning binary representations of instances and classes is a classical problem with several high potential applications. In modern settings, the compression of high-dimensional neural representations to low-dimensional binary codes is a challenging task and often require large bit-codes to be accurate. In this work, we propose a novel method for Learning Low-dimensional binary Codes (LLC) for instances as well as classes. Our method does not require any side-information, like annotated attributes or label meta-data, and learns extremely low-dimensional binary codes (~20 bits for ImageNet-1K). The learnt codes are super-efficient while still ensuring nearly optimal classification accuracy for ResNet50 on ImageNet-1K. We demonstrate that the learnt codes capture intrinsically important features in the data, by discovering an intuitive taxonomy over classes. We further quantitatively measure the quality of our codes by applying it to the efficient image retrieval as well as out-of-distribution (OOD) detection problems. For ImageNet-100 retrieval problem, our learnt binary codes outperform 16 bit HashNet using only 10 bits and also are as accurate as 10 dimensional real representations. Finally, our learnt binary codes can perform OOD detection, out-of-the-box, as accurately as a baseline that needs ~3000 samples to tune its threshold, while we require none. Code and pre-trained models are available at this https URL." @default.
• W3171091521 created "2021-06-22" @default.
• W3171091521 creator A5016535318 @default.
• W3171091521 creator A5018792915 @default.
• W3171091521 creator A5026171268 @default.
• W3171091521 creator A5030680279 @default.
• W3171091521 creator A5034432097 @default.
• W3171091521 creator A5035949823 @default.
• W3171091521 creator A5058266558 @default.
• W3171091521 creator A5059282528 @default.
• W3171091521 creator A5060895155 @default.
• W3171091521 date "2021-06-02" @default.
• W3171091521 modified "2023-09-23" @default.
• W3171091521 title "LLC: Accurate, Multi-purpose Learnt Low-dimensional Binary Codes" @default.
• W3171091521 cites W1676820704 @default.
• W3171091521 cites W1940008012 @default.
• W3171091521 cites W1943357268 @default.
• W3171091521 cites W1953590900 @default.
• W3171091521 cites W1974647172 @default.
• W3171091521 cites W1978962787 @default.
• W3171091521 cites W205159212 @default.
• W3171091521 cites W2081580037 @default.
• W3171091521 cites W2098411764 @default.
• W3171091521 cites W21006490 @default.
• W3171091521 cites W2101276256 @default.
• W3171091521 cites W2108598243 @default.
• W3171091521 cites W2113307832 @default.
• W3171091521 cites W2125560515 @default.
• W3171091521 cites W2128532956 @default.
• W3171091521 cites W2128715378 @default.
• W3171091521 cites W2134270519 @default.
• W3171091521 cites W2144892774 @default.
• W3171091521 cites W2154109204 @default.
• W3171091521 cites W2155144632 @default.
• W3171091521 cites W2157791002 @default.
• W3171091521 cites W2158703410 @default.
• W3171091521 cites W2161449253 @default.
• W3171091521 cites W2162006472 @default.
• W3171091521 cites W2163605009 @default.
• W3171091521 cites W2164338181 @default.
• W3171091521 cites W2171061940 @default.
• W3171091521 cites W2171790913 @default.
• W3171091521 cites W2173418170 @default.
• W3171091521 cites W2183087644 @default.
• W3171091521 cites W2194775991 @default.
• W3171091521 cites W2242818861 @default.
• W3171091521 cites W2267635276 @default.
• W3171091521 cites W2292376821 @default.
• W3171091521 cites W2300242332 @default.
• W3171091521 cites W2464915613 @default.
• W3171091521 cites W2531327146 @default.
• W3171091521 cites W2536185904 @default.
• W3171091521 cites W2612445135 @default.
• W3171091521 cites W2781821509 @default.
• W3171091521 cites W2890960987 @default.
• W3171091521 cites W2897765688 @default.
• W3171091521 cites W2906963924 @default.
• W3171091521 cites W2913932916 @default.
• W3171091521 cites W2951238624 @default.
• W3171091521 cites W2963213486 @default.
• W3171091521 cites W2963469388 @default.
• W3171091521 cites W2964280870 @default.
• W3171091521 cites W2970449868 @default.
• W3171091521 cites W2970967090 @default.
• W3171091521 cites W3009806461 @default.
• W3171091521 cites W3034724020 @default.
• W3171091521 cites W3035304835 @default.
• W3171091521 cites W3035305184 @default.
• W3171091521 cites W3102124616 @default.
• W3171091521 cites W3114079967 @default.
• W3171091521 cites W3209770953 @default.
• W3171091521 cites W93016980 @default.
• W3171091521 hasPublicationYear "2021" @default.
• W3171091521 type Work @default.
• W3171091521 sameAs 3171091521 @default.
• W3171091521 citedByCount "0" @default.
• W3171091521 crossrefType "posted-content" @default.
• W3171091521 hasAuthorship W3171091521A5016535318 @default.
• W3171091521 hasAuthorship W3171091521A5018792915 @default.
• W3171091521 hasAuthorship W3171091521A5026171268 @default.
• W3171091521 hasAuthorship W3171091521A5030680279 @default.
• W3171091521 hasAuthorship W3171091521A5034432097 @default.
• W3171091521 hasAuthorship W3171091521A5035949823 @default.
• W3171091521 hasAuthorship W3171091521A5058266558 @default.
• W3171091521 hasAuthorship W3171091521A5059282528 @default.
• W3171091521 hasAuthorship W3171091521A5060895155 @default.
• W3171091521 hasConcept C11413529 @default.
• W3171091521 hasConcept C119857082 @default.
• W3171091521 hasConcept C124101348 @default.
• W3171091521 hasConcept C153180895 @default.
• W3171091521 hasConcept C154945302 @default.
• W3171091521 hasConcept C177264268 @default.
• W3171091521 hasConcept C199360897 @default.
• W3171091521 hasConcept C2776760102 @default.
• W3171091521 hasConcept C2779190172 @default.
• W3171091521 hasConcept C2780009758 @default.
• W3171091521 hasConcept C33923547 @default.
• W3171091521 hasConcept C41008148 @default.
• W3171091521 hasConcept C43126263 @default.
• W3171091521 hasConcept C48372109 @default.

• next