DTSemNet gives an exact, invertible neural-network encoding of hard oblique decision trees that supports direct gradient training for both classification and regression without probabilistic softening or quantized estimators.
Vanilla gradient descent for oblique decision trees
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
citation-role summary
other 1
citation-polarity summary
fields
cs.LG 2years
2026 2verdicts
UNVERDICTED 2roles
other 1polarities
unclear 1representative citing papers
HRT reframes oblique splits as Newton-optimized hinge regressions achieving universal approximation with O(δ²) rate, and HRT-Boost ensembles them with stage-wise empirical risk reduction guarantees under squared loss.
citing papers explorer
-
Approximation-Free Differentiable Oblique Decision Trees
DTSemNet gives an exact, invertible neural-network encoding of hard oblique decision trees that supports direct gradient training for both classification and regression without probabilistic softening or quantized estimators.
-
Hinge Regression Trees and HRT-Boost: Newton-Optimized Oblique Learning for Compact Tabular Models
HRT reframes oblique splits as Newton-optimized hinge regressions achieving universal approximation with O(δ²) rate, and HRT-Boost ensembles them with stage-wise empirical risk reduction guarantees under squared loss.