A new combination of mixture-of-experts, multi-head attention, and a linear orthogonal regularizer reduces individual treatment effect estimation error on IHDP, ACIC 2016, Jobs, and a large-scale production dataset.
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
-
Deep Disentangled Representation Network for Treatment Effect Estimation
A new combination of mixture-of-experts, multi-head attention, and a linear orthogonal regularizer reduces individual treatment effect estimation error on IHDP, ACIC 2016, Jobs, and a large-scale production dataset.