English (United Kingdom)

https://curated-unify.zendy.io/wp-json/zendy-region/v1/featured_content/oa?rat=en

https://curated-unify.zendy.io/wp-json/zendy-region/v1/highlighted_journal/

Global: Zendy Plus annual

Presents the access of premium content as premium feature

Premium Content

Presents the keyphrase highlighting as premium feature

Keyphrase Highlighting

Presents the summarisation as premium feature

Summarisation

Insights

Presents the pdf analysis as premium feature

PDF Analysis

Presents the zaia usage as premium feature

ZAIA

Global: Zendy Plus monthly

Global: Zendy Tools annual

Global: Zendy Tools monthly

Global: Zendy Free Plan

Evolutionary Computation algorithms have been used to solve optimizationproblems in relation with architectural, hyper-parameter or trainingconfiguration, forging the field known today as Neural Architecture Search.These algorithms have been combined with other techniques such as the pruningof Neural Networks, which reduces the complexity of the network, and theTransfer Learning, which lets the import of knowledge from another problemrelated to the one at hand. The usage of several criteria to evaluate thequality of the evolutionary proposals is also a common case, in which theperformance and complexity of the network are the most used criteria. This workproposes MO-EvoPruneDeepTL, a multi-objective evolutionary pruning algorithm.\proposal uses Transfer Learning to adapt the last layers of Deep NeuralNetworks, by replacing them with sparse layers evolved by a genetic algorithm,which guides the evolution based in the performance, complexity and robustnessof the network, being the robustness a great quality indicator for the evolvedmodels. We carry out different experiments with several datasets to assess thebenefits of our proposal. Results show that our proposal achieves promisingresults in all the objectives, and direct relation are presented among them.The experiments also show that the most influential neurons help us explainwhich parts of the input images are the most relevant for the prediction of thepruned neural network. Lastly, by virtue of the diversity within the Paretofront of pruning patterns produced by the proposal, it is shown that anensemble of differently pruned models improves the overall performance androbustness of the trained networks.

Multiobjective Evolutionary Pruning of Deep Neural Networks with Transfer Learning for improving their Performance and Robustness