FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W1676755748> ?p ?o ?g. }

• W1676755748 abstract "Deep learning is a popular field that encompasses a range of multi-layer connectionist techniques. While these techniques have achieved great success on a number of difficult computer vision problems, the representation biases that allow this success have not been thoroughly explored. In this paper, we examine the hypothesis that one strength of many deep learning algorithms is their ability to exploit spatially local statistical information. We present a formal description of how data vectors can be partitioned into sub-vectors that preserve spatially local information. As a test case, we then use statistical models to examine how much of such structure exists in the MNIST dataset. Finally, we present experimental results from training RBMs using partitioned data, and demonstrate the advantages they have over non-partitioned RBMs. Through these results, we show how the performance advantage is reliant on spatially local structure, by demonstrating the performance impact of randomly permuting the input data to destroy local structure. Overall, our results support the hypothesis that a representation bias reliant upon spatially local statistical information can improve performance, so long as this bias is a good match for the data. We also suggest statistical tools for determining a priori whether a dataset is a good match for this bias or not." @default.
• W1676755748 created "2016-06-24" @default.
• W1676755748 creator A5015261502 @default.
• W1676755748 creator A5052140189 @default.
• W1676755748 creator A5072522101 @default.
• W1676755748 date "2015-07-01" @default.
• W1676755748 modified "2023-09-25" @default.
• W1676755748 title "Deep learning using partitioned data vectors" @default.
• W1676755748 cites W140273596 @default.
• W1676755748 cites W1813659000 @default.
• W1676755748 cites W1964155876 @default.
• W1676755748 cites W1986369746 @default.
• W1676755748 cites W1993709822 @default.
• W1676755748 cites W1995341919 @default.
• W1676755748 cites W2032657310 @default.
• W1676755748 cites W2072128103 @default.
• W1676755748 cites W2099866409 @default.
• W1676755748 cites W2100495367 @default.
• W1676755748 cites W2101926813 @default.
• W1676755748 cites W2107878631 @default.
• W1676755748 cites W2110798204 @default.
• W1676755748 cites W2112796928 @default.
• W1676755748 cites W2116064496 @default.
• W1676755748 cites W2116825644 @default.
• W1676755748 cites W2123925574 @default.
• W1676755748 cites W2134557905 @default.
• W1676755748 cites W2136922672 @default.
• W1676755748 cites W2137388705 @default.
• W1676755748 cites W2138806976 @default.
• W1676755748 cites W2138857742 @default.
• W1676755748 cites W2143022286 @default.
• W1676755748 cites W2146337213 @default.
• W1676755748 cites W2203408497 @default.
• W1676755748 cites W2326628520 @default.
• W1676755748 cites W2344372818 @default.
• W1676755748 cites W2606321545 @default.
• W1676755748 cites W2950500591 @default.
• W1676755748 cites W2962771908 @default.
• W1676755748 cites W36434594 @default.
• W1676755748 cites W593140685 @default.
• W1676755748 doi "https://doi.org/10.1109/ijcnn.2015.7280484" @default.
• W1676755748 hasPublicationYear "2015" @default.
• W1676755748 type Work @default.
• W1676755748 sameAs 1676755748 @default.
• W1676755748 citedByCount "7" @default.
• W1676755748 countsByYear W16767557482016 @default.
• W1676755748 countsByYear W16767557482019 @default.
• W1676755748 countsByYear W16767557482020 @default.
• W1676755748 countsByYear W16767557482023 @default.
• W1676755748 crossrefType "proceedings-article" @default.
• W1676755748 hasAuthorship W1676755748A5015261502 @default.
• W1676755748 hasAuthorship W1676755748A5052140189 @default.
• W1676755748 hasAuthorship W1676755748A5072522101 @default.
• W1676755748 hasBestOaLocation W16767557482 @default.
• W1676755748 hasConcept C105795698 @default.
• W1676755748 hasConcept C108583219 @default.
• W1676755748 hasConcept C111472728 @default.
• W1676755748 hasConcept C114289077 @default.
• W1676755748 hasConcept C119857082 @default.
• W1676755748 hasConcept C124101348 @default.
• W1676755748 hasConcept C138885662 @default.
• W1676755748 hasConcept C154945302 @default.
• W1676755748 hasConcept C165696696 @default.
• W1676755748 hasConcept C17744445 @default.
• W1676755748 hasConcept C190502265 @default.
• W1676755748 hasConcept C199539241 @default.
• W1676755748 hasConcept C202444582 @default.
• W1676755748 hasConcept C2776359362 @default.
• W1676755748 hasConcept C33923547 @default.
• W1676755748 hasConcept C38652104 @default.
• W1676755748 hasConcept C41008148 @default.
• W1676755748 hasConcept C75553542 @default.
• W1676755748 hasConcept C80444323 @default.
• W1676755748 hasConcept C87007009 @default.
• W1676755748 hasConcept C94625758 @default.
• W1676755748 hasConcept C9652623 @default.
• W1676755748 hasConceptScore W1676755748C105795698 @default.
• W1676755748 hasConceptScore W1676755748C108583219 @default.
• W1676755748 hasConceptScore W1676755748C111472728 @default.
• W1676755748 hasConceptScore W1676755748C114289077 @default.
• W1676755748 hasConceptScore W1676755748C119857082 @default.
• W1676755748 hasConceptScore W1676755748C124101348 @default.
• W1676755748 hasConceptScore W1676755748C138885662 @default.
• W1676755748 hasConceptScore W1676755748C154945302 @default.
• W1676755748 hasConceptScore W1676755748C165696696 @default.
• W1676755748 hasConceptScore W1676755748C17744445 @default.
• W1676755748 hasConceptScore W1676755748C190502265 @default.
• W1676755748 hasConceptScore W1676755748C199539241 @default.
• W1676755748 hasConceptScore W1676755748C202444582 @default.
• W1676755748 hasConceptScore W1676755748C2776359362 @default.
• W1676755748 hasConceptScore W1676755748C33923547 @default.
• W1676755748 hasConceptScore W1676755748C38652104 @default.
• W1676755748 hasConceptScore W1676755748C41008148 @default.
• W1676755748 hasConceptScore W1676755748C75553542 @default.
• W1676755748 hasConceptScore W1676755748C80444323 @default.
• W1676755748 hasConceptScore W1676755748C87007009 @default.
• W1676755748 hasConceptScore W1676755748C94625758 @default.
• W1676755748 hasConceptScore W1676755748C9652623 @default.
• W1676755748 hasLocation W16767557481 @default.
• W1676755748 hasLocation W16767557482 @default.

• next