AttributionScanner: A Visual Analytics System for Model Validation with Metadata-Free Slice Finding

arXiv:2401.06462v2 Announce Type: replace
Abstract: Data slice finding is an emerging technique for validating machine learning (ML) models by identifying and analyzing subgroups in a dataset that exhibit poor performance, often characterized by distinct feature sets or descriptive metadata. However, in the context of validating vision models involving unstructured image data, this approach faces significant challenges, including the laborious and costly requirement for additional metadata and the complex task of interpreting the root causes of underperformance. To address these challenges, …

abstract analytics arxiv context cs.cv cs.hc data dataset feature free however machine machine learning metadata performance slice subgroups type validation vision vision models visual visual analytics