Matches in SemOpenAlex for { <https://semopenalex.org/work/W2979391681> ?p ?o ?g. }
- W2979391681 abstract "Continual learning models allow to learn and adapt to new changes and tasks over time. However, in continual and sequential learning scenarios in which the models are trained using different data with various distributions, neural networks tend to forget the previously learned knowledge. This phenomenon is often referred to as catastrophic forgetting. The catastrophic forgetting is an inevitable problem in continual learning models for dynamic environments. To address this issue, we propose a method, called Continual Bayesian Learning Networks (CBLN), which enables the networks to allocate additional resources to adapt to new tasks without forgetting the previously learned tasks. Using a Bayesian Neural Network, CBLN maintains a mixture of Gaussian posterior distributions that are associated with different tasks. The proposed method tries to optimise the number of resources that are needed to learn each task and avoids an exponential increase in the number of resources that are involved in learning multiple tasks. The proposed method does not need to access the past training data and can choose suitable weights to classify the data points during the test time automatically based on an uncertainty criterion. We have evaluated our method on the MNIST and UCR time-series datasets. The evaluation results show that our method can address the catastrophic forgetting problem at a promising rate compared to the state-of-the-art models." @default.
- W2979391681 created "2019-10-18" @default.
- W2979391681 creator A5041622837 @default.
- W2979391681 creator A5056157061 @default.
- W2979391681 creator A5058205038 @default.
- W2979391681 creator A5090478829 @default.
- W2979391681 date "2019-10-09" @default.
- W2979391681 modified "2023-09-24" @default.
- W2979391681 title "Continual Learning Using Bayesian Neural Networks" @default.
- W2979391681 cites W1546503963 @default.
- W2979391681 cites W1682403713 @default.
- W2979391681 cites W1799366690 @default.
- W2979391681 cites W1904826605 @default.
- W2979391681 cites W1985690171 @default.
- W2979391681 cites W2037979274 @default.
- W2979391681 cites W2038705219 @default.
- W2979391681 cites W2047057213 @default.
- W2979391681 cites W2077723394 @default.
- W2979391681 cites W2094192576 @default.
- W2979391681 cites W2095705004 @default.
- W2979391681 cites W2103753221 @default.
- W2979391681 cites W2108598243 @default.
- W2979391681 cites W2113839990 @default.
- W2979391681 cites W2116522068 @default.
- W2979391681 cites W2122838776 @default.
- W2979391681 cites W2122882636 @default.
- W2979391681 cites W2124776405 @default.
- W2979391681 cites W2138309709 @default.
- W2979391681 cites W2194775991 @default.
- W2979391681 cites W2560647685 @default.
- W2979391681 cites W2600383743 @default.
- W2979391681 cites W2605043629 @default.
- W2979391681 cites W2737492962 @default.
- W2979391681 cites W2743151379 @default.
- W2979391681 cites W2744654301 @default.
- W2979391681 cites W2765101016 @default.
- W2979391681 cites W2765895026 @default.
- W2979391681 cites W2770864858 @default.
- W2979391681 cites W2788388592 @default.
- W2979391681 cites W2939618326 @default.
- W2979391681 cites W2951266961 @default.
- W2979391681 cites W2963559848 @default.
- W2979391681 cites W2964917234 @default.
- W2979391681 cites W2974317861 @default.
- W2979391681 cites W3030364939 @default.
- W2979391681 doi "https://doi.org/10.48550/arxiv.1910.04112" @default.
- W2979391681 hasPublicationYear "2019" @default.
- W2979391681 type Work @default.
- W2979391681 sameAs 2979391681 @default.
- W2979391681 citedByCount "2" @default.
- W2979391681 countsByYear W29793916812020 @default.
- W2979391681 countsByYear W29793916812021 @default.
- W2979391681 crossrefType "posted-content" @default.
- W2979391681 hasAuthorship W2979391681A5041622837 @default.
- W2979391681 hasAuthorship W2979391681A5056157061 @default.
- W2979391681 hasAuthorship W2979391681A5058205038 @default.
- W2979391681 hasAuthorship W2979391681A5090478829 @default.
- W2979391681 hasBestOaLocation W29793916811 @default.
- W2979391681 hasConcept C107673813 @default.
- W2979391681 hasConcept C119857082 @default.
- W2979391681 hasConcept C127413603 @default.
- W2979391681 hasConcept C138885662 @default.
- W2979391681 hasConcept C154945302 @default.
- W2979391681 hasConcept C190502265 @default.
- W2979391681 hasConcept C201995342 @default.
- W2979391681 hasConcept C2780451532 @default.
- W2979391681 hasConcept C33724603 @default.
- W2979391681 hasConcept C41008148 @default.
- W2979391681 hasConcept C41895202 @default.
- W2979391681 hasConcept C50644808 @default.
- W2979391681 hasConcept C7149132 @default.
- W2979391681 hasConceptScore W2979391681C107673813 @default.
- W2979391681 hasConceptScore W2979391681C119857082 @default.
- W2979391681 hasConceptScore W2979391681C127413603 @default.
- W2979391681 hasConceptScore W2979391681C138885662 @default.
- W2979391681 hasConceptScore W2979391681C154945302 @default.
- W2979391681 hasConceptScore W2979391681C190502265 @default.
- W2979391681 hasConceptScore W2979391681C201995342 @default.
- W2979391681 hasConceptScore W2979391681C2780451532 @default.
- W2979391681 hasConceptScore W2979391681C33724603 @default.
- W2979391681 hasConceptScore W2979391681C41008148 @default.
- W2979391681 hasConceptScore W2979391681C41895202 @default.
- W2979391681 hasConceptScore W2979391681C50644808 @default.
- W2979391681 hasConceptScore W2979391681C7149132 @default.
- W2979391681 hasLocation W29793916811 @default.
- W2979391681 hasLocation W29793916812 @default.
- W2979391681 hasOpenAccess W2979391681 @default.
- W2979391681 hasPrimaryLocation W29793916811 @default.
- W2979391681 hasRelatedWork W2786958491 @default.
- W2979391681 hasRelatedWork W2904174853 @default.
- W2979391681 hasRelatedWork W2944572476 @default.
- W2979391681 hasRelatedWork W2978290780 @default.
- W2979391681 hasRelatedWork W2979391681 @default.
- W2979391681 hasRelatedWork W3082871683 @default.
- W2979391681 hasRelatedWork W3126776133 @default.
- W2979391681 hasRelatedWork W4287234591 @default.
- W2979391681 hasRelatedWork W4312120628 @default.
- W2979391681 hasRelatedWork W1629725936 @default.
- W2979391681 isParatext "false" @default.
- W2979391681 isRetracted "false" @default.