Paired autoencoders with linear latent-space maps, interpreted through Bayes risk minimization, give theory and experiments for inverse problems and beat an end-to-end baseline when paired training data are scarce.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
A Paired Autoencoder Framework for Inverse Problems via Bayes Risk Minimization
Paired autoencoders with linear latent-space maps, interpreted through Bayes risk minimization, give theory and experiments for inverse problems and beat an end-to-end baseline when paired training data are scarce.