Name
Abstract | ||
|---|---|---|
In recent years, the multiclass and mutlilabel classification problems we encounter in many applications have very large ($10^3$–$10^6$) number of classes. However, each instance belongs to only one or few classes, i.e., the label vectors are sparse. In this work, we propose a novel approach based on group testing to solve such large multilabel classification problems with sparse label vectors. We describe various group testing constructions, and advocate the use of concatenated Reed Solomon codes and unbalanced bipartite expander graphs for extreme classification problems. The proposed approach has several advantages theoretically and practically over existing popular methods. Our method operates on the binary alphabet and can utilize the well-established binary classifiers for learning. The error correction capabilities of the codes are leveraged for the first time in the learning problem to correct prediction errors. Even if a linearly growing number of classifiers mis-classify, these errors are fully corrected. We establish Hamming loss error bounds for the approach. More importantly, our method utilizes a simple prediction algorithm and does not require matrix inversion or solving optimization problems making the algorithm very inexpensive. Numerical experiments with various datasets illustrate the superior performance of our method. |
Year | Venue | Field |
|---|---|---|
2017 | ICML | Computer science,Artificial intelligence,Group testing,Machine learning |
DocType | Citations | PageRank |
Conference | 1 | 0.34 |
References | Authors | |
0 | 2 | |
Name | Order | Citations | PageRank |
|---|---|---|---|
| shashanka ubaru | 1 | 58 | 8.97 |
| Arya Mazumdar | 2 | 307 | 41.81 |