BLAE adds injective regularization via a separation criterion and bi-Lipschitz constraints to guarantee injectivity and geometric preservation in autoencoders, outperforming prior methods on manifold fidelity under sparsity and distribution shifts.
Parametric umap embeddings for represen- tation and semisupervised learning.Neural Computation, 33(11):2881–2907
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2026 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Bi-Lipschitz Autoencoder With Injectivity Guarantee
BLAE adds injective regularization via a separation criterion and bi-Lipschitz constraints to guarantee injectivity and geometric preservation in autoencoders, outperforming prior methods on manifold fidelity under sparsity and distribution shifts.