Matches in SemOpenAlex for { <https://semopenalex.org/work/W2963686698> ?p ?o ?g. }
Showing items 1 to 77 of
77
with 100 items per page.
- W2963686698 endingPage "1702" @default.
- W2963686698 startingPage "1696" @default.
- W2963686698 abstract "In this paper, we focus on online representation learning in non-stationary environments which may require continuous adaptation of model architecture. We propose a novel online dictionary-learning (sparse-coding) framework which incorporates the addition and deletion of hidden units (dictionary elements), and is inspired by the adult neurogenesis phenomenon in the dentate gyrus of the hippocampus, known to be associated with improved cognitive function and adaptation to new environments. In the online learning setting, where new input instances arrive sequentially in batches, the neuronal-birth is implemented by adding new units with random initial weights (random dictionary elements); the number of new units is determined by the current performance (representation error) of the dictionary, higher error causing an increase in the birth rate. Neuronal-death is implemented by imposing l1/l2-regularization (group sparsity) on the dictionary within the block-coordinate descent optimization at each iteration of our online alternating minimization scheme, which iterates between the code and dictionary updates. Finally, hidden unit connectivity adaptation is facilitated by introducing sparsity in dictionary elements. Our empirical evaluation on several real-life datasets (images and language) as well as on synthetic data demonstrates that the proposed approach can considerably outperform the state-of-art fixed-size (nonadaptive) online sparse coding of Mairal et al. (2009) in the presence of nonstationary data. Moreover, we identify certain properties of the data (e.g., sparse inputs with nearly non-overlapping supports) and of the model (e.g., dictionary sparsity) associated with such improvements." @default.
- W2963686698 created "2019-07-30" @default.
- W2963686698 creator A5005877741 @default.
- W2963686698 creator A5055054658 @default.
- W2963686698 creator A5055430458 @default.
- W2963686698 creator A5065396462 @default.
- W2963686698 date "2017-02-16" @default.
- W2963686698 modified "2023-10-04" @default.
- W2963686698 title "Neurogenesis-Inspired Dictionary Learning: Online Model Adaption in a Changing World." @default.
- W2963686698 hasPublicationYear "2017" @default.
- W2963686698 type Work @default.
- W2963686698 sameAs 2963686698 @default.
- W2963686698 citedByCount "0" @default.
- W2963686698 crossrefType "proceedings-article" @default.
- W2963686698 hasAuthorship W2963686698A5005877741 @default.
- W2963686698 hasAuthorship W2963686698A5055054658 @default.
- W2963686698 hasAuthorship W2963686698A5055430458 @default.
- W2963686698 hasAuthorship W2963686698A5065396462 @default.
- W2963686698 hasConcept C105795698 @default.
- W2963686698 hasConcept C119857082 @default.
- W2963686698 hasConcept C124066611 @default.
- W2963686698 hasConcept C137293760 @default.
- W2963686698 hasConcept C153180895 @default.
- W2963686698 hasConcept C154771677 @default.
- W2963686698 hasConcept C154945302 @default.
- W2963686698 hasConcept C157553263 @default.
- W2963686698 hasConcept C2777851325 @default.
- W2963686698 hasConcept C2988886741 @default.
- W2963686698 hasConcept C33923547 @default.
- W2963686698 hasConcept C41008148 @default.
- W2963686698 hasConcept C59404180 @default.
- W2963686698 hasConcept C77637269 @default.
- W2963686698 hasConcept C80444323 @default.
- W2963686698 hasConceptScore W2963686698C105795698 @default.
- W2963686698 hasConceptScore W2963686698C119857082 @default.
- W2963686698 hasConceptScore W2963686698C124066611 @default.
- W2963686698 hasConceptScore W2963686698C137293760 @default.
- W2963686698 hasConceptScore W2963686698C153180895 @default.
- W2963686698 hasConceptScore W2963686698C154771677 @default.
- W2963686698 hasConceptScore W2963686698C154945302 @default.
- W2963686698 hasConceptScore W2963686698C157553263 @default.
- W2963686698 hasConceptScore W2963686698C2777851325 @default.
- W2963686698 hasConceptScore W2963686698C2988886741 @default.
- W2963686698 hasConceptScore W2963686698C33923547 @default.
- W2963686698 hasConceptScore W2963686698C41008148 @default.
- W2963686698 hasConceptScore W2963686698C59404180 @default.
- W2963686698 hasConceptScore W2963686698C77637269 @default.
- W2963686698 hasConceptScore W2963686698C80444323 @default.
- W2963686698 hasLocation W29636866981 @default.
- W2963686698 hasOpenAccess W2963686698 @default.
- W2963686698 hasPrimaryLocation W29636866981 @default.
- W2963686698 hasRelatedWork W1483455099 @default.
- W2963686698 hasRelatedWork W1674801529 @default.
- W2963686698 hasRelatedWork W209548358 @default.
- W2963686698 hasRelatedWork W2102075063 @default.
- W2963686698 hasRelatedWork W2176058272 @default.
- W2963686698 hasRelatedWork W2180695702 @default.
- W2963686698 hasRelatedWork W2185231601 @default.
- W2963686698 hasRelatedWork W2198221632 @default.
- W2963686698 hasRelatedWork W2262760538 @default.
- W2963686698 hasRelatedWork W2285179135 @default.
- W2963686698 hasRelatedWork W2564717828 @default.
- W2963686698 hasRelatedWork W2583093513 @default.
- W2963686698 hasRelatedWork W2595951978 @default.
- W2963686698 hasRelatedWork W2771297143 @default.
- W2963686698 hasRelatedWork W2780938819 @default.
- W2963686698 hasRelatedWork W2792944346 @default.
- W2963686698 hasRelatedWork W2953275220 @default.
- W2963686698 hasRelatedWork W3025553125 @default.
- W2963686698 hasRelatedWork W3207930698 @default.
- W2963686698 hasRelatedWork W888734479 @default.
- W2963686698 isParatext "false" @default.
- W2963686698 isRetracted "false" @default.
- W2963686698 magId "2963686698" @default.
- W2963686698 workType "article" @default.