MINIMAX RATE OF DISTRIBUTION ESTIMATION ON UNKNOWN SUBMANIFOLDS UNDER ADVERSARIAL LOSSES

Keyphrases

• Distribution Estimation 100%
• Minimax Rate 100%
• Adversarial Loss 100%
• Minimax 50%
• Low-dimensional Manifolds 50%
• Target Distribution 50%
• Manifold Structure 50%
• Modeling Approach 25%
• Fundamental Limits 25%
• Large Collection 25%
• Low-dimensional Structures 25%
• Generative Models 25%
• Problem Features 25%
• High-dimensional Data 25%
• Statistical Inference 25%
• Machine Learning 25%
• Adaptive Estimator 25%
• Partition of Unity 25%
• Distribution Modelling 25%
• Underlying Distribution 25%
• Optimal Rate 25%
• Differential Geometry 25%
• Intrinsic Dimension 25%
• Global Parametrization 25%
• Distribution Class 25%
• Latent Feature 25%
• Data Manifold 25%
• Latent Representation 25%
• Smoothness Assumption 25%
• Generative Modeling 25%
• High-dimensional Distribution 25%
• Deep Generative Modeling 25%
• Information Decoding 25%

Computer Science

• Dimensional Manifold 100%
• Target Distribution 100%
• Machine Learning 50%
• Dimensional Structure 50%
• Fundamental Limit 50%
• Parametrization 50%
• Generative Model 50%
• Underlying Data 50%
• High Dimensional Data 50%
• Class Distribution 50%
• Underlying Distribution 50%
• Intrinsic Dimensionality 50%
• Smoothness Assumption 50%
• Dimensional Distribution 50%

Mathematics

• Minimax 100%
• Submanifold 100%
• Manifold 75%
• Dimensional Manifold 50%
• target distribution π 50%
• Modeling Approach 25%
• Upper Bound 25%
• Dimensional Structure 25%
• Inferential Statistics 25%
• Partition of Unity 25%
• Underlying Distribution 25%
• Differential Geometry 25%
• Dimensional Distribution 25%
• Dimensional Data 25%
• Deep Learning 25%