Augment & Valuate : A Data Enhancement Pipeline for Data-Centric AI
This addresses data quality issues for industrial ML applications, but it is incremental as it combines existing methods into a pipeline.
The paper tackled data scarcity and noise in industrial ML by proposing a domain-agnostic pipeline for refining data quality in image classification, achieving 84.711% test accuracy and ranking #6 in the Data-Centric AI competition.
Data scarcity and noise are important issues in industrial applications of machine learning. However, it is often challenging to devise a scalable and generalized approach to address the fundamental distributional and semantic properties of dataset with black box models. For this reason, data-centric approaches are crucial for the automation of machine learning operation pipeline. In order to serve as the basis for this automation, we suggest a domain-agnostic pipeline for refining the quality of data in image classification problems. This pipeline contains data valuation, cleansing, and augmentation. With an appropriate combination of these methods, we could achieve 84.711% test accuracy (ranked #6, Honorable Mention in the Most Innovative) in the Data-Centric AI competition only with the provided dataset.