Extreme Multi-label Classification from Aggregated Labels
arXiv:2004.00198
Abstract
Extreme multi-label classification (XMC) is the problem of finding the relevant labels for an input, from a very large universe of possible labels. We consider XMC in the setting where labels are available only for groups of samples - but not for individual ones. Current XMC approaches are not built for such multi-instance multi-label (MIML) training data, and MIML approaches do not scale to XMC sizes. We develop a new and scalable algorithm to impute individual-sample labels from the group labels; this can be paired with any existing XMC method to solve the aggregated label problem. We characterize the statistical properties of our algorithm under mild assumptions, and provide a new end-to-end framework for MIML as an extension. Experiments on both aggregated label XMC and MIML tasks show the advantages over existing approaches.
References in corpus (13)
- Classification with Noisy Labels by Importance Reweighting
- Multi-Instance Multi-Label Learning
- Large-scale Multi-label Learning with Missing Labels
- LSHTC: A Benchmark for Large-Scale Text Classification
- Learning with Bad Training Data via Iterative Trimmed Loss Minimization
- Linear regression without correspondence
- Linear Regression with Shuffled Labels
- Signal Recovery from Unlabeled Samples
- Minimax rates in permutation estimation for feature matching
- Extreme Classification in Log Memory using Count-Min Sketch: A Case Study of Amazon Search with 50M Products
- Taming Pretrained Transformers for Extreme Multi-label Text Classification
- Deep Probabilistic Logic: A Unifying Framework for Indirect Supervision
- HAXMLNet: Hierarchical Attention Network for Extreme Multi-Label Text Classification