Learning Label Embeddings for Nearest-Neighbor Multi-class Classification with an Application to Speech Recognition
We consider the problem of using nearest neighbor methods to provide a conditional probability estimate, P(y|a), when the number of labels y is large and the labels share some underlying structure. We propose a method for learning label embeddings (similar to error-correcting output codes (ECOCs)) t...
| Main Authors: | , |
|---|---|
| Format: | Article |
| Language: | English |
| Published: |
Neural Information Processing Systems (NIPS) Foundation,
2010-10-14T18:09:55Z.
|
| Subjects: | |
| Online Access: | Get fulltext |