Universal regular conditional distributions via probabilistic transformers

DOI10.1007/s00365-023-09635-3zbMath1529.68277arXiv2105.07743OpenAlexW4361018823MaRDI QIDQ6101232

Anastasis Kratsios

Publication date: 20 June 2023

Published in: Constructive Approximation (Search for Journal in Brave)

Full work available at URL: https://arxiv.org/abs/2105.07743

zbMATH Keywords

universal approximationtransformersgeometric deep learningcomputational optimal transportregular conditional distributionsmeasure-valued neural networks

Mathematics Subject Classification ID

Artificial neural networks and deep learning (68T07) Probability distributions: general theory (60E05) Approximation by other special function classes (41A30) Integration and disintegration of measures (28A50) Optimal transportation (49Q22)

Related Items

Generative OrnsteinUhlenbeck markets via geometric deep learning

Cites Work

• Bi-Lipschitz approximation by finite-dimensional imbeddings
• On embedding uniform and topological spaces
• Filling Riemannian manifolds
• Fractional Brownian motion and the Markov property
• Lectures on analysis on metric spaces
• Assouad's theorem with dimension independent of the snowflaking
• Approximation rates for neural networks with general activation functions
• Convergence of stochastic proximal gradient algorithm
• The universal approximation property. Characterization, construction, representation, and existence
• Optimal approximation rate of ReLU networks in terms of width and depth
• Approximation spaces of deep neural networks
• Rejoinder: ``Nonparametric regression using deep neural networks with ReLU activation function
• Universality of deep convolutional neural networks
• Hardy spaces on Ahlfors-regular quasi metric spaces. A sharp theory
• A note on generalized inverses
• On topological infinite deficiency
• Quasisymmetric dimension distortion of Ahlfors regular subsets of a metric space
• On the structure of Lipschitz-free spaces
• Approximation by finitely supported measures
• Metaphors in systolic geometry
• Semmes surfaces and intrinsic Lipschitz graphs in the Heisenberg group
• Better Bootstrap Confidence Intervals
• Autoregressive Conditional Heteroscedasticity with Estimates of the Variance of United Kingdom Inflation
• Colloquium lectures on geometric measure theory
• Universal approximation bounds for superpositions of a sigmoidal function
• On the unique predual problem for Lipschitz spaces
• Uniform decomposition of probability measures: quantization, clustering and rate of convergence
• Lipschitz Algebras
• Lipschitz-free Banach spaces
• A GEOMETRIC STUDY OF WASSERSTEIN SPACES: HADAMARD SPACES
• An L¹ ergodic theorem with values in a non-positively curved space via a canonical barycenter map
• Stochastic Calculus for Fractional Brownian Motion I. Theory
• Deep Neural Networks, Generic Universal Interpolation, and Controlled ODEs
• Deep Network Approximation for Smooth Functions
• Equivalence of approximation by convolutional neural networks and fully-connected networks
• An Invitation to Statistics in Wasserstein Space
• Regularization and Variable Selection Via the Elastic Net
• Introduction to Stochastic Analysis and Malliavin Calculus
• Understanding Machine Learning
• Book Review: Optimal transport: old and new
• Ridge Regression: Biased Estimation for Nonorthogonal Problems
• A logical calculus of the ideas immanent in nervous activity
• Probability theory. A comprehensive course
• Convex analysis and monotone operator theory in Hilbert spaces
• Approximation by superpositions of a sigmoidal function
• Locally flat imbeddings of topological manifolds
• Stochastic gradient boosting.
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item