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

While deep neural networks have demonstrated remarkable performance acrossvarious tasks, they typically require massive training data. Due to thepresence of redundancies and biases in real-world datasets, not all data in thetraining dataset contributes to the model performance. To address this issue,dataset pruning techniques have been introduced to enhance model performanceand efficiency by eliminating redundant training samples and reducingcomputational and memory overhead. However, previous works most rely onmanually crafted scalar scores, limiting their practical performance andscalability across diverse deep networks and datasets. In this paper, wepropose AdaPruner, an end-to-end Adaptive DAtaset PRUNing framEwoRk. AdaPrunercan perform effective dataset pruning without the need for explicitly definedmetrics. Our framework jointly prunes training data and fine-tunes models withtask-specific optimization objectives. AdaPruner leverages (1) An adaptivedataset pruning (ADP) module, which iteratively prunes redundant samples to anexpected pruning ratio; and (2) A pruning performance controller (PPC) module,which optimizes the model performance for accurate pruning. Therefore,AdaPruner exhibits high scalability and compatibility across various datasetsand deep networks, yielding improved dataset distribution and enhanced modelperformance. AdaPruner can still significantly enhance model performance evenafter pruning up to 10-30\% of the training data. Notably, these improvementsare accompanied by substantial savings in memory and computation costs.Qualitative and quantitative experiments suggest that AdaPruner outperformsother state-of-the-art dataset pruning methods by a large margin.

Not All Data Matters: An End-to-End Adaptive Dataset Pruning Framework for Enhancing Model Performance and Efficiency