Matches in SemOpenAlex for { <https://semopenalex.org/work/W4313229768> ?p ?o ?g. }
- W4313229768 endingPage "2874" @default.
- W4313229768 startingPage "2862" @default.
- W4313229768 abstract "In this work, we introduce a learning model designed to meet the needs of applications in which computational resources are limited, and robustness and interpretability are prioritized. Learning problems can be formulated as constrained stochastic optimization problems, with the constraints originating mainly from model assumptions that define a tradeoff between complexity and performance. This tradeoff is closely related to overfitting, generalization capacity, and robustness to noise and adversarial attacks, and depends on both the structure and complexity of the model, as well as the properties of the optimization methods used. We develop an online prototype-based learning algorithm based on annealing optimization that is formulated as an online gradient-free stochastic approximation algorithm. The learning model can be viewed as an interpretable and progressively growing competitive-learning neural network model to be used for supervised, unsupervised, and reinforcement learning. The annealing nature of the algorithm contributes to minimal hyperparameter tuning requirements, poor local minima prevention, and robustness with respect to the initial conditions. At the same time, it provides online control over the performance–complexity tradeoff by progressively increasing the complexity of the learning model as needed, through an intuitive bifurcation phenomenon. Finally, the use of stochastic approximation enables the study of the convergence of the learning algorithm through mathematical tools from dynamical systems and control, and allows for its integration with reinforcement learning algorithms, constructing an adaptive state–action aggregation scheme." @default.
- W4313229768 created "2023-01-06" @default.
- W4313229768 creator A5014738965 @default.
- W4313229768 creator A5044395526 @default.
- W4313229768 date "2023-05-01" @default.
- W4313229768 modified "2023-09-29" @default.
- W4313229768 title "Annealing Optimization for Progressive Learning With Stochastic Approximation" @default.
- W4313229768 cites W1498436455 @default.
- W4313229768 cites W1892947258 @default.
- W4313229768 cites W1994616650 @default.
- W4313229768 cites W2032558547 @default.
- W4313229768 cites W2071983464 @default.
- W4313229768 cites W2091565802 @default.
- W4313229768 cites W2094364653 @default.
- W4313229768 cites W2136637968 @default.
- W4313229768 cites W2136922672 @default.
- W4313229768 cites W2161877964 @default.
- W4313229768 cites W2170615774 @default.
- W4313229768 cites W2272817333 @default.
- W4313229768 cites W2919115771 @default.
- W4313229768 cites W2963809228 @default.
- W4313229768 cites W2963857521 @default.
- W4313229768 cites W2983041926 @default.
- W4313229768 cites W3100573453 @default.
- W4313229768 cites W3128191354 @default.
- W4313229768 cites W3154762694 @default.
- W4313229768 cites W3186158026 @default.
- W4313229768 cites W32403112 @default.
- W4313229768 cites W4210298964 @default.
- W4313229768 cites W4210453992 @default.
- W4313229768 cites W4233696721 @default.
- W4313229768 cites W4289829119 @default.
- W4313229768 cites W4315488697 @default.
- W4313229768 doi "https://doi.org/10.1109/tac.2022.3232706" @default.
- W4313229768 hasPublicationYear "2023" @default.
- W4313229768 type Work @default.
- W4313229768 citedByCount "2" @default.
- W4313229768 countsByYear W43132297682022 @default.
- W4313229768 countsByYear W43132297682023 @default.
- W4313229768 crossrefType "journal-article" @default.
- W4313229768 hasAuthorship W4313229768A5014738965 @default.
- W4313229768 hasAuthorship W4313229768A5044395526 @default.
- W4313229768 hasBestOaLocation W43132297682 @default.
- W4313229768 hasConcept C104317684 @default.
- W4313229768 hasConcept C11413529 @default.
- W4313229768 hasConcept C115903097 @default.
- W4313229768 hasConcept C119857082 @default.
- W4313229768 hasConcept C126255220 @default.
- W4313229768 hasConcept C134306372 @default.
- W4313229768 hasConcept C137836250 @default.
- W4313229768 hasConcept C154945302 @default.
- W4313229768 hasConcept C185592680 @default.
- W4313229768 hasConcept C186633575 @default.
- W4313229768 hasConcept C194387892 @default.
- W4313229768 hasConcept C26517878 @default.
- W4313229768 hasConcept C33923547 @default.
- W4313229768 hasConcept C38652104 @default.
- W4313229768 hasConcept C41008148 @default.
- W4313229768 hasConcept C50644808 @default.
- W4313229768 hasConcept C55479107 @default.
- W4313229768 hasConcept C55493867 @default.
- W4313229768 hasConcept C63479239 @default.
- W4313229768 hasConcept C91873725 @default.
- W4313229768 hasConcept C97541855 @default.
- W4313229768 hasConceptScore W4313229768C104317684 @default.
- W4313229768 hasConceptScore W4313229768C11413529 @default.
- W4313229768 hasConceptScore W4313229768C115903097 @default.
- W4313229768 hasConceptScore W4313229768C119857082 @default.
- W4313229768 hasConceptScore W4313229768C126255220 @default.
- W4313229768 hasConceptScore W4313229768C134306372 @default.
- W4313229768 hasConceptScore W4313229768C137836250 @default.
- W4313229768 hasConceptScore W4313229768C154945302 @default.
- W4313229768 hasConceptScore W4313229768C185592680 @default.
- W4313229768 hasConceptScore W4313229768C186633575 @default.
- W4313229768 hasConceptScore W4313229768C194387892 @default.
- W4313229768 hasConceptScore W4313229768C26517878 @default.
- W4313229768 hasConceptScore W4313229768C33923547 @default.
- W4313229768 hasConceptScore W4313229768C38652104 @default.
- W4313229768 hasConceptScore W4313229768C41008148 @default.
- W4313229768 hasConceptScore W4313229768C50644808 @default.
- W4313229768 hasConceptScore W4313229768C55479107 @default.
- W4313229768 hasConceptScore W4313229768C55493867 @default.
- W4313229768 hasConceptScore W4313229768C63479239 @default.
- W4313229768 hasConceptScore W4313229768C91873725 @default.
- W4313229768 hasConceptScore W4313229768C97541855 @default.
- W4313229768 hasFunder F4320332180 @default.
- W4313229768 hasIssue "5" @default.
- W4313229768 hasLocation W43132297681 @default.
- W4313229768 hasLocation W43132297682 @default.
- W4313229768 hasOpenAccess W4313229768 @default.
- W4313229768 hasPrimaryLocation W43132297681 @default.
- W4313229768 hasRelatedWork W1561685851 @default.
- W4313229768 hasRelatedWork W2090984822 @default.
- W4313229768 hasRelatedWork W2122994540 @default.
- W4313229768 hasRelatedWork W2283831177 @default.
- W4313229768 hasRelatedWork W2576759595 @default.
- W4313229768 hasRelatedWork W2785308959 @default.
- W4313229768 hasRelatedWork W2913526350 @default.