Universal Approximation of Measure-to-Measure Operators by Pushforwards
cs.LG, cs.NA, math.NA, stat.ML
Submitted: 2026-09-28
Updated: 2026-09-28
Comments: 31 pages (9 main text, 19 appendix, and 3 references pages)
License: http://arxiv.org/licenses/nonexclusive-distrib/1.0/
Terminology
Sources
- Learning Probabilistic Filters with Strictly Proper Scoring Rules
- Denseness of adapted processes among causal couplings
- In-Context Operator Learning on the Space of Probability Measures
- On the Expressive Power of Transformers for Contextual Relations
- Transformers through the lens of support-preserving maps between measures
- Continuous transformations of probability measures and their transport representations
- One Operator for Many Densities: Amortized Approximation of Conditioning by Neural Operators
- A Mathematical Theory of Attention
- On the Regularity of Attention
- Learning Likelihoods with Conditional Normalizing Flows
Related papers
- Polynomial-Augmented Neural Networks (PANNs) with Weak Orthogonality Constraints for Enhanced Function and PDE Approximation
- AIRL-S: Unifying Reinforcement Learning and Search-Based Test-Time Scaling via Adversarial Inverse Reinforcement Learning
- Transformers as Bayesian In-Context Experimenters: Smoothness-Adaptive Efficient ATE Estimation
- Convergence issues in Relational Concept Analysis based on AOC-posets
- Beliefs Beyond Posteriors: The Predictive Cost of ELBO Binding
- Understanding Diffusion Models via Ratio-Based Function Approximation with SignReLU Networks