A two-step deep non-negative autoencoder method for extreme multi-label learning yields both predictions and interpretable label hierarchies and dependencies.
As discussed previously, our proposed non-negative autoencoder is a kind of generalization of the NMF and its non-negative conceptual label sets are relatively easy to interpret
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2019 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Towards Interpretable Deep Extreme Multi-label Learning
A two-step deep non-negative autoencoder method for extreme multi-label learning yields both predictions and interpretable label hierarchies and dependencies.