Derives near-optimal nonasymptotic excess-risk bounds for Engression and reverse Markov Engression over Hölder classes via energy distance.
arXiv preprint arXiv:2502.02483 , year=
4 Pith papers cite this work. Polarity classification is still indexing.
4
Pith papers citing it
years
2026 4verdicts
UNVERDICTED 4representative citing papers
Augments the energy score objective for sample-based generative models with a differentiable decision loss that is itself a proper scoring rule, yielding targeted improvements on cost-sensitive regions in synthetic and real tasks.
Establishes convergence guarantees for overparameterized 2-layer ReLU networks in flow matching, generalization bounds for the velocity-field objective, and Wasserstein guarantees for generated samples, using multi-task representation learning bounds.
The book presents principles from optimization and information theory to explain deep network architectures and enable new interpretable models.
citing papers explorer
-
Theoretical Analysis of Engression and Reverse Markov Engression
Derives near-optimal nonasymptotic excess-risk bounds for Engression and reverse Markov Engression over Hölder classes via energy distance.
-
Decision-Aware Training for Sample-Based Generative Models
Augments the energy score objective for sample-based generative models with a differentiable decision loss that is itself a proper scoring rule, yielding targeted improvements on cost-sensitive regions in synthetic and real tasks.
-
A Theory on Flow Matching with Neural Networks
Establishes convergence guarantees for overparameterized 2-layer ReLU networks in flow matching, generalization bounds for the velocity-field objective, and Wasserstein guarantees for generated samples, using multi-task representation learning bounds.
-
Principles and Practice of Deep Representation Learning: or a Mathematical Theory of Memory
The book presents principles from optimization and information theory to explain deep network architectures and enable new interpretable models.