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

Background: Embedded feature selection in high-dimensional data with verysmall sample sizes requires optimized hyperparameters for the model buildingprocess. For this hyperparameter optimization, nested cross-validation must beapplied to avoid a biased performance estimation. The resulting repeatedtraining with high-dimensional data leads to very long computation times.Moreover, it is likely to observe a high variance in the individual performanceevaluation metrics caused by outliers in tiny validation sets. Therefore, earlystopping applying standard pruning algorithms to save time risks discardingpromising hyperparameter sets.  Result: To speed up feature selection for high-dimensional data with tinysample size, we adapt the use of a state-of-the-art asynchronous successivehalving pruner. In addition, we combine it with two complementary pruningstrategies based on domain or prior knowledge. One pruning strategy immediatelystops computing trials with semantically meaningless results for the selectedhyperparameter combinations. The other is a new extrapolating threshold pruningstrategy suitable for nested-cross-validation with a high variance ofperformance evaluation metrics. In repeated experiments, our combined pruningstrategy keeps all promising trials. At the same time, the calculation time issubstantially reduced compared to using a state-of-the-art asynchronoussuccessive halving pruner alone. Up to 81.3\% fewer models were trainedachieving the same optimization result.  Conclusion: The proposed combined pruning strategy accelerates data analysisor enables deeper searches for hyperparameters within the same computationtime. This leads to significant savings in time, money and energy consumption,opening the door to advanced, time-consuming analyses.

Combined Pruning for Nested Cross-Validation to Accelerate Automated  Hyperparameter Optimization for Embedded Feature Selection in  High-Dimensional Data with Very Small Sample Sizes