Matches in SemOpenAlex for { <https://semopenalex.org/work/W3002177189> ?p ?o ?g. }
- W3002177189 abstract "In a conventional supervised learning setting, a machine learning model has access to examples of all object classes that are desired to be recognized during the inference stage. This results in a fixed model that lacks the flexibility to adapt to new learning tasks. In practical settings, learning tasks often arrive in a sequence and the models must continually learn to increment their previously acquired knowledge. Existing incremental learning approaches fall well below the state-of-the-art cumulative models that use all training classes at once. In this paper, we propose a random path selection algorithm, called Adaptive RPS-Net, that progressively chooses optimal paths for the new tasks while encouraging parameter sharing between tasks. We introduce a new network capacity measure that enables us to automatically switch paths if the already used resources are saturated. Since the proposed path-reuse strategy ensures forward knowledge transfer, our approach is efficient and has considerably less computation overhead. As an added novelty, the proposed model integrates knowledge distillation and retrospection along with the path selection strategy to overcome catastrophic forgetting. In order to maintain an equilibrium between previous and newly acquired knowledge, we propose a simple controller to dynamically balance the model plasticity. Through extensive experiments, we demonstrate that the Adaptive RPS-Net method surpasses the state-of-the-art performance for incremental learning and by utilizing parallel computation this method can run in constant time with nearly the same efficiency as a conventional deep convolutional neural network." @default.
- W3002177189 created "2020-01-30" @default.
- W3002177189 creator A5000300751 @default.
- W3002177189 creator A5041437150 @default.
- W3002177189 creator A5053765366 @default.
- W3002177189 creator A5063179713 @default.
- W3002177189 creator A5079112987 @default.
- W3002177189 creator A5082634513 @default.
- W3002177189 date "2019-06-03" @default.
- W3002177189 modified "2023-09-25" @default.
- W3002177189 title "An Adaptive Random Path Selection Approach for Incremental Learning" @default.
- W3002177189 cites W1522301498 @default.
- W3002177189 cites W157468466 @default.
- W3002177189 cites W1682403713 @default.
- W3002177189 cites W1821462560 @default.
- W3002177189 cites W2143386621 @default.
- W3002177189 cites W2160684493 @default.
- W3002177189 cites W2194775991 @default.
- W3002177189 cites W2401932923 @default.
- W3002177189 cites W2473930607 @default.
- W3002177189 cites W2560647685 @default.
- W3002177189 cites W2583761661 @default.
- W3002177189 cites W2737492962 @default.
- W3002177189 cites W2765895026 @default.
- W3002177189 cites W2785366763 @default.
- W3002177189 cites W2788388592 @default.
- W3002177189 cites W2795061970 @default.
- W3002177189 cites W2884282566 @default.
- W3002177189 cites W2894094671 @default.
- W3002177189 cites W2895723011 @default.
- W3002177189 cites W2899063268 @default.
- W3002177189 cites W2921184280 @default.
- W3002177189 cites W2929125140 @default.
- W3002177189 cites W2948336998 @default.
- W3002177189 cites W2962724315 @default.
- W3002177189 cites W2963038864 @default.
- W3002177189 cites W2963559848 @default.
- W3002177189 cites W2963560049 @default.
- W3002177189 cites W2963588172 @default.
- W3002177189 cites W2963588253 @default.
- W3002177189 cites W2963850662 @default.
- W3002177189 cites W2964081807 @default.
- W3002177189 cites W2964189064 @default.
- W3002177189 cites W2981864462 @default.
- W3002177189 cites W2995139074 @default.
- W3002177189 cites W2426267443 @default.
- W3002177189 hasPublicationYear "2019" @default.
- W3002177189 type Work @default.
- W3002177189 sameAs 3002177189 @default.
- W3002177189 citedByCount "4" @default.
- W3002177189 countsByYear W30021771892020 @default.
- W3002177189 countsByYear W30021771892021 @default.
- W3002177189 crossrefType "posted-content" @default.
- W3002177189 hasAuthorship W3002177189A5000300751 @default.
- W3002177189 hasAuthorship W3002177189A5041437150 @default.
- W3002177189 hasAuthorship W3002177189A5053765366 @default.
- W3002177189 hasAuthorship W3002177189A5063179713 @default.
- W3002177189 hasAuthorship W3002177189A5079112987 @default.
- W3002177189 hasAuthorship W3002177189A5082634513 @default.
- W3002177189 hasConcept C105795698 @default.
- W3002177189 hasConcept C111919701 @default.
- W3002177189 hasConcept C119857082 @default.
- W3002177189 hasConcept C154945302 @default.
- W3002177189 hasConcept C199360897 @default.
- W3002177189 hasConcept C2776214188 @default.
- W3002177189 hasConcept C2777735758 @default.
- W3002177189 hasConcept C2779960059 @default.
- W3002177189 hasConcept C2780598303 @default.
- W3002177189 hasConcept C33923547 @default.
- W3002177189 hasConcept C41008148 @default.
- W3002177189 hasConceptScore W3002177189C105795698 @default.
- W3002177189 hasConceptScore W3002177189C111919701 @default.
- W3002177189 hasConceptScore W3002177189C119857082 @default.
- W3002177189 hasConceptScore W3002177189C154945302 @default.
- W3002177189 hasConceptScore W3002177189C199360897 @default.
- W3002177189 hasConceptScore W3002177189C2776214188 @default.
- W3002177189 hasConceptScore W3002177189C2777735758 @default.
- W3002177189 hasConceptScore W3002177189C2779960059 @default.
- W3002177189 hasConceptScore W3002177189C2780598303 @default.
- W3002177189 hasConceptScore W3002177189C33923547 @default.
- W3002177189 hasConceptScore W3002177189C41008148 @default.
- W3002177189 hasLocation W30021771891 @default.
- W3002177189 hasOpenAccess W3002177189 @default.
- W3002177189 hasPrimaryLocation W30021771891 @default.
- W3002177189 hasRelatedWork W1988431231 @default.
- W3002177189 hasRelatedWork W2111770102 @default.
- W3002177189 hasRelatedWork W2225266074 @default.
- W3002177189 hasRelatedWork W2884778159 @default.
- W3002177189 hasRelatedWork W2892978054 @default.
- W3002177189 hasRelatedWork W2898290017 @default.
- W3002177189 hasRelatedWork W2906904540 @default.
- W3002177189 hasRelatedWork W2972740417 @default.
- W3002177189 hasRelatedWork W2998093422 @default.
- W3002177189 hasRelatedWork W3008124487 @default.
- W3002177189 hasRelatedWork W3030598573 @default.
- W3002177189 hasRelatedWork W3034867292 @default.
- W3002177189 hasRelatedWork W3035486480 @default.
- W3002177189 hasRelatedWork W3082320588 @default.
- W3002177189 hasRelatedWork W3091795298 @default.
- W3002177189 hasRelatedWork W3109062428 @default.