A new adaptive ℓ₂-regularized Newton boosting algorithm for decision trees delivers global O(1/k²) convergence on general convex losses, recovering classical Newton boosting as a special case under stronger assumptions.
2016 IEEE Conference on Computer Vision and Pattern Recognition (CVPR) , year=
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
years
2026 4verdicts
UNVERDICTED 4representative citing papers
Introduces a fairness layer for deep learning models that guarantees output parity and an online primal-dual algorithm for aggregate fairness guarantees in streaming predictions with small batch sizes.
StableGrad applies scale correction to weight gradients after backpropagation to enable stable optimization of deep BatchNorm-free networks including PINNs.
A graph-spectral importance score based on layer-wise structural distortion between pre- and post-activation neuron graphs identifies removable neurons for iterative pruning without intermediate updates, followed by recovery fine-tuning.
citing papers explorer
-
Gradient Regularized Newton Boosting Trees with Global Convergence
A new adaptive ℓ₂-regularized Newton boosting algorithm for decision trees delivers global O(1/k²) convergence on general convex losses, recovering classical Newton boosting as a special case under stronger assumptions.
-
Differentiable Optimization Layers for Guaranteed Fairness in Deep Learning
Introduces a fairness layer for deep learning models that guarantees output parity and an online primal-dual algorithm for aggregate fairness guarantees in streaming predictions with small batch sizes.
-
StableGrad: Backward Scale Control without Batch Normalization
StableGrad applies scale correction to weight gradients after backpropagation to enable stable optimization of deep BatchNorm-free networks including PINNs.
-
Spectral structural distortion reveals redundant neurons in neural networks
A graph-spectral importance score based on layer-wise structural distortion between pre- and post-activation neuron graphs identifies removable neurons for iterative pruning without intermediate updates, followed by recovery fine-tuning.