FRINK Linked Data Fragments server

Matches in SemOpenAlex for { <https://semopenalex.org/work/W2950149565> ?p ?o ?g. }

• W2950149565 abstract "Deep neural networks and decision trees operate on largely separate paradigms; typically, the former performs representation learning with pre-specified architectures, while the latter is characterised by learning hierarchies over pre-specified features with data-driven architectures. We unite the two via adaptive neural trees (ANTs) that incorporates representation learning into edges, routing functions and leaf nodes of a decision tree, along with a backpropagation-based training algorithm that adaptively grows the architecture from primitive modules (e.g., convolutional layers). We demonstrate that, whilst achieving competitive performance on classification and regression datasets, ANTs benefit from (i) lightweight inference via conditional computation, (ii) hierarchical separation of features useful to the task e.g. learning meaningful class associations, such as separating natural vs. man-made objects, and (iii) a mechanism to adapt the architecture to the size and complexity of the training dataset." @default.
• W2950149565 created "2019-06-27" @default.
• W2950149565 creator A5004486211 @default.
• W2950149565 creator A5040477021 @default.
• W2950149565 creator A5064157732 @default.
• W2950149565 creator A5065210301 @default.
• W2950149565 creator A5091150085 @default.
• W2950149565 date "2018-07-17" @default.
• W2950149565 modified "2023-09-27" @default.
• W2950149565 title "Adaptive Neural Trees" @default.
• W2950149565 cites W137456267 @default.
• W2950149565 cites W148877464 @default.
• W2950149565 cites W1522301498 @default.
• W2950149565 cites W1549825062 @default.
• W2950149565 cites W1568207135 @default.
• W2950149565 cites W1613249581 @default.
• W2950149565 cites W1665214252 @default.
• W2950149565 cites W1678356000 @default.
• W2950149565 cites W1849277567 @default.
• W2950149565 cites W1855112655 @default.
• W2950149565 cites W2025653905 @default.
• W2950149565 cites W2059424674 @default.
• W2950149565 cites W2081136789 @default.
• W2950149565 cites W2097117768 @default.
• W2950149565 cites W2098458263 @default.
• W2950149565 cites W2107285841 @default.
• W2950149565 cites W2108598243 @default.
• W2950149565 cites W2109779438 @default.
• W2950149565 cites W2111935653 @default.
• W2950149565 cites W2112796928 @default.
• W2950149565 cites W2123199310 @default.
• W2950149565 cites W2135293965 @default.
• W2950149565 cites W2136922672 @default.
• W2950149565 cites W2156163116 @default.
• W2950149565 cites W2179423374 @default.
• W2950149565 cites W2184406811 @default.
• W2950149565 cites W2194775991 @default.
• W2950149565 cites W2220384803 @default.
• W2950149565 cites W2242818861 @default.
• W2950149565 cites W2325237720 @default.
• W2950149565 cites W2520615899 @default.
• W2950149565 cites W2530601379 @default.
• W2950149565 cites W2553303224 @default.
• W2950149565 cites W2562731582 @default.
• W2950149565 cites W2592340788 @default.
• W2950149565 cites W2622604310 @default.
• W2950149565 cites W2744654301 @default.
• W2950149565 cites W2748513770 @default.
• W2950149565 cites W2755458934 @default.
• W2950149565 cites W2769421449 @default.
• W2950149565 cites W2774627531 @default.
• W2950149565 cites W2780170571 @default.
• W2950149565 cites W2899771611 @default.
• W2950149565 cites W2911964244 @default.
• W2950149565 cites W2949264490 @default.
• W2950149565 cites W2962677625 @default.
• W2950149565 cites W2962707453 @default.
• W2950149565 cites W2962917663 @default.
• W2950149565 cites W2963175113 @default.
• W2950149565 cites W2963382180 @default.
• W2950149565 cites W2963446712 @default.
• W2950149565 cites W2963518064 @default.
• W2950149565 cites W2963536136 @default.
• W2950149565 cites W2963628712 @default.
• W2950149565 cites W2963703618 @default.
• W2950149565 cites W2963793535 @default.
• W2950149565 cites W2963993763 @default.
• W2950149565 cites W2964091842 @default.
• W2950149565 cites W2965435631 @default.
• W2950149565 cites W3118608800 @default.
• W2950149565 hasPublicationYear "2018" @default.
• W2950149565 type Work @default.
• W2950149565 sameAs 2950149565 @default.
• W2950149565 citedByCount "27" @default.
• W2950149565 countsByYear W29501495652018 @default.
• W2950149565 countsByYear W29501495652019 @default.
• W2950149565 countsByYear W29501495652020 @default.
• W2950149565 countsByYear W29501495652021 @default.
• W2950149565 crossrefType "posted-content" @default.
• W2950149565 hasAuthorship W2950149565A5004486211 @default.
• W2950149565 hasAuthorship W2950149565A5040477021 @default.
• W2950149565 hasAuthorship W2950149565A5064157732 @default.
• W2950149565 hasAuthorship W2950149565A5065210301 @default.
• W2950149565 hasAuthorship W2950149565A5091150085 @default.
• W2950149565 hasConcept C108583219 @default.
• W2950149565 hasConcept C113174947 @default.
• W2950149565 hasConcept C119857082 @default.
• W2950149565 hasConcept C134306372 @default.
• W2950149565 hasConcept C154945302 @default.
• W2950149565 hasConcept C155032097 @default.
• W2950149565 hasConcept C17744445 @default.
• W2950149565 hasConcept C199539241 @default.
• W2950149565 hasConcept C2776214188 @default.
• W2950149565 hasConcept C2776359362 @default.
• W2950149565 hasConcept C2777212361 @default.
• W2950149565 hasConcept C33923547 @default.
• W2950149565 hasConcept C41008148 @default.
• W2950149565 hasConcept C50644808 @default.
• W2950149565 hasConcept C81363708 @default.
• W2950149565 hasConcept C84525736 @default.

• next