Learning Likelihoods with Conditional Normalizing Flows
arXiv:1912.00042
Abstract
Normalizing Flows (NFs) are able to model complicated distributions p(y) with strong inter-dimensional correlations and high multimodality by transforming a simple base density p(z) through an invertible neural network under the change of variables formula. Such behavior is desirable in multivariate structured prediction tasks, where handcrafted per-pixel loss-based methods inadequately capture strong correlations between output dimensions. We present a study of conditional normalizing flows (CNFs), a class of NFs where the base density to output space mapping is conditioned on an input x, to model conditional densities p(y|x). CNFs are efficient in sampling and inference, they can be trained with a likelihood-based objective, and CNFs, being generative flows, do not suffer from mode collapse or training instabilities. We provide an effective method to train continuous CNFs for binary problems and in particular, we apply these CNFs to super-resolution and vessel segmentation tasks demonstrating competitive performance on standard benchmark datasets in terms of likelihood and conventional metrics.
18 pages, 8 Tables, 9 Figures, Preprint
References in corpus (3)
Cited by in corpus (20)
- Normalizing Flows for Probabilistic Modeling and Inference
- i-flow: High-dimensional Integration and Sampling with Normalizing Flows
- Deep Structural Causal Models for Tractable Counterfactual Inference
- Argmax Flows and Multinomial Diffusion: Learning Categorical Distributions
- Wavelet Flow: Fast Training of High Resolution Normalizing Flows
- Learning Discrete Distributions by Dequantization
- Faster Uncertainty Quantification for Inverse Problems with Conditional Normalizing Flows
- Multimodal Safety-Critical Scenarios Generation for Decision-Making Algorithms Evaluation
- Pseudo-Rehearsal for Continual Learning with Normalizing Flows
- Learning to Plan Optimally with Flow-based Motion Planner
- Conditional Normalizing Flows for Low-Dose Computed Tomography Image Reconstruction
- Generative modeling of time-dependent densities via optimal transport and projection pursuit
- Featurized Density Ratio Estimation
- An information-based metric for observing strategy optimization, demonstrated in the context of photometric redshifts with applications to cosmology
- Scalable Safety-Critical Policy Evaluation with Accelerated Rare Event Sampling
- Noise Conditional Flow Model for Learning the Super-Resolution Space
- Prior Flow Variational Autoencoder: A density estimation model for Non-Intrusive Load Monitoring
- Deep Measurement Updates for Bayes Filters
- Viscos Flows: Variational Schur Conditional Sampling With Normalizing Flows
- Generating Data Augmentation samples for Semantic Segmentation of Salt Bodies in a Synthetic Seismic Image Dataset