Neural Relation Graph: A Unified Framework for Identifying Label Noise and Outlier Data
About
Diagnosing and cleaning data is a crucial step for building robust machine learning systems. However, identifying problems within large-scale datasets with real-world distributions is challenging due to the presence of complex issues such as label errors, under-representation, and outliers. In this paper, we propose a unified approach for identifying the problematic data by utilizing a largely ignored source of information: a relational structure of data in the feature-embedded space. To this end, we present scalable and effective algorithms for detecting label errors and outlier data based on the relational graph structure of data. We further introduce a visualization tool that provides contextual information of a data point in the feature-embedded space, serving as an effective tool for interactively diagnosing data. We evaluate the label error and outlier/out-of-distribution (OOD) detection performances of our approach on the large-scale image, speech, and language domain tasks, including ImageNet, ESC-50, and SST2. Our approach achieves state-of-the-art detection performance on all tasks considered and demonstrates its effectiveness in debugging large-scale real-world datasets across various domains. We release codes at https://github.com/snu-mllab/Neural-Relation-Graph.
Related benchmarks
| Task | Dataset | Result | Rank | |
|---|---|---|---|---|
| Distribution Shift Detection | BROAD (test) | Novel Classes AUC93.61 | 40 | |
| Error detection | Adversarial Attacks (test) | AUC79.1 | 40 | |
| Error detection | Average All shifts (test) | AUC83.4 | 40 | |
| OOD Detection | CoComageNet | Detection AUC0.6479 | 40 | |
| Error detection | In-distribution (test) | AUC0.7691 | 40 | |
| Error detection | Corruptions (test) | AUC95 | 40 | |
| OOD Detection | CoComageNet mono | Detection AUC0.5226 | 40 |