Density estimation using Real NVP

Abstract

Unsupervised learning of probabilistic models is a central yet challenging problem in machine learning. Specifically, designing models with tractable learning, sampling, inference and evaluation is crucial in solving this task. We extend the space of such models using real-valued non-volume preserving (real NVP) transformations, a set of powerful invertible and learnable transformations, resulting in an unsupervised learning algorithm with exact log-likelihood computation, exact sampling, exact inference of latent variables, and an interpretable latent space. We demonstrate its ability to model natural images on four datasets through sampling, log-likelihood evaluation and latent variable manipulations.

10 pages of main content, 3 pages of bibliography, 18 pages of appendix. Accepted at ICLR 2017

References in corpus (8)

• Very Deep Convolutional Networks for Large-Scale Image Recognition
• TensorFlow: Large-Scale Machine Learning on Heterogeneous Distributed Systems
• Deep Generative Image Models using a Laplacian Pyramid of Adversarial Networks
• NICE: Non-linear Independent Components Estimation
• Resnet in Resnet: Generalizing Residual Architectures
• Markov Chain Monte Carlo and Variational Inference: Bridging the Gap
• MADE: Masked Autoencoder for Distribution Estimation
• Improving Variational Inference with Inverse Autoregressive Flow

Cited by in corpus (463)

• Denoising Diffusion Probabilistic Models
• An Introduction to Variational Autoencoders
• Conditional Image Synthesis With Auxiliary Classifier GANs
• NIPS 2016 Tutorial: Generative Adversarial Networks
• Score-Based Generative Modeling through Stochastic Differential Equations
• Physics-Constrained Deep Learning for High-dimensional Surrogate Modeling and Uncertainty Quantification without Labeled Data
• What Makes for Good Views for Contrastive Learning?
• Energy-based Out-of-distribution Detection
• NVAE: A Deep Hierarchical Variational Autoencoder
• Modeling the Dynamics of PDE Systems with Physics-Constrained Deep Auto-Regressive Networks
• On Mutual Information Maximization for Representation Learning
• Guided Image Generation with Conditional Invertible Neural Networks
• FastFlow: Unsupervised Anomaly Detection and Localization via 2D Normalizing Flows
• A Connection between Generative Adversarial Networks, Inverse Reinforcement Learning, and Energy-Based Models
• GraphAF: a Flow-based Autoregressive Model for Molecular Graph Generation
• Masked Autoregressive Flow for Density Estimation
• The Reversible Residual Network: Backpropagation Without Storing Activations
• A Survey on Neural Speech Synthesis
• Glow: Generative Flow with Invertible 1x1 Convolutions
• Improving Variational Inference with Inverse Autoregressive Flow
• Mining gold from implicit models to improve likelihood-free inference
• Pre-training Molecular Graph Representation with 3D Geometry
• Getting High: High Fidelity Simulation of High Granularity Calorimeters with High Speed
• The Fishyscapes Benchmark: Measuring Blind Spots in Semantic Segmentation
• WINNet: Wavelet-inspired Invertible Network for Image Denoising
• VideoGPT: Video Generation using VQ-VAE and Transformers
• Classification Accuracy Score for Conditional Generative Models
• Nested Sampling with Normalising Flows for Gravitational-Wave Inference
• Sharp Minima Can Generalize For Deep Nets
• BIVA: A Very Deep Hierarchy of Latent Variables for Generative Modeling
• Sylvester Normalizing Flows for Variational Inference
• i-flow: High-dimensional Integration and Sampling with Normalizing Flows
• Generating Diverse High-Fidelity Images with VQ-VAE-2
• Simple and Principled Uncertainty Estimation with Deterministic Deep Learning via Distance Awareness
• Denoising Diffusion Implicit Models
• Roundtrip: A Deep Generative Neural Density Estimator
• How to train your neural ODE: the world of Jacobian and kinetic regularization
• Learning and Evaluating Representations for Deep One-class Classification
• AI Feynman 2.0: Pareto-optimal symbolic regression exploiting graph modularity
• Nonparametric Density Estimation for High-Dimensional Data - Algorithms and Applications
• Solving inverse problems using conditional invertible neural networks
• Structured Denoising Diffusion Models in Discrete State-Spaces
• How to Train Your Energy-Based Models
• Flowtron: an Autoregressive Flow-based Generative Network for Text-to-Speech Synthesis
• Understanding Event-Generation Networks via Uncertainties
• Inference Suboptimality in Variational Autoencoders
• ANODE: Unconditionally Accurate Memory-Efficient Gradients for Neural ODEs
• Mining for Dark Matter Substructure: Inferring subhalo population properties from strong lenses with machine learning
• SATAR: A Self-supervised Approach to Twitter Account Representation Learning and its Application in Bot Detection
• Bayesian Layers: A Module for Neural Network Uncertainty
• PixelCNN++: Improving the PixelCNN with Discretized Logistic Mixture Likelihood and Other Modifications
• Reversible Architectures for Arbitrarily Deep Residual Neural Networks
• Deep Generative Modelling: A Comparative Review of VAEs, GANs, Normalizing Flows, Energy-Based and Autoregressive Models
• Generative Networks for Precision Enthusiasts
• Equivariant Flows: Exact Likelihood Generative Learning for Symmetric Densities
• Latent Space Policies for Hierarchical Reinforcement Learning
• Contracting Arbitrary Tensor Networks: General Approximate Algorithm and Applications in Graphical Models and Quantum Circuit Simulations
• A review of radar-based nowcasting of precipitation and applicable machine learning techniques
• A Survey of Deep Learning for Scientific Discovery
• i-RevNet: Deep Invertible Networks
• A Perspective on Deep Learning for Molecular Modeling and Simulations
• Deep Imitative Models for Flexible Inference, Planning, and Control
• Augmented Neural ODEs
• PixelSNAIL: An Improved Autoregressive Generative Model
• Learning Independent Features with Adversarial Nets for Non-linear ICA
• Adaptive deep density approximation for Fokker-Planck equations
• SoftFlow: Probabilistic Framework for Normalizing Flow on Manifolds
• Disentanglement by Nonlinear ICA with General Incompressible-flow Networks (GIN)
• Learning identifiable and interpretable latent models of high-dimensional neural activity using pi-VAE
• Is Generator Conditioning Causally Related to GAN Performance?
• Parallel Multiscale Autoregressive Density Estimation
• Distribution Matching in Variational Inference
• Very Deep VAEs Generalize Autoregressive Models and Can Outperform Them on Images
• Equivariant Flows: sampling configurations for multi-body systems with symmetric energies
• Autoregressive Denoising Diffusion Models for Multivariate Probabilistic Time Series Forecasting
• Generative adversarial interpolative autoencoding: adversarial training on latent space interpolations encourage convex latent distributions
• The LHC Olympics 2020: A Community Challenge for Anomaly Detection in High Energy Physics
• Invertible generative models for inverse problems: mitigating representation error and dataset bias
• Comparison of Maximum Likelihood and GAN-based training of Real NVPs
• Coupling-based Invertible Neural Networks Are Universal Diffeomorphism Approximators
• Equivariant Hamiltonian Flows
• Graph Normalizing Flows
• Dynamical mass inference of galaxy clusters with neural flows
• Inference of cosmic-ray source properties by conditional invertible neural networks
• IDF++: Analyzing and Improving Integer Discrete Flows for Lossless Compression
• Can multi-label classification networks know what they don't know?
• Automated curricula through setter-solver interactions
• Understanding Anomaly Detection with Deep Invertible Networks through Hierarchies of Distributions and Features
• Inductive Representation Learning on Temporal Graphs
• Policy Gradient based Quantum Approximate Optimization Algorithm
• Stellar Parameter Determination from Photometry using Invertible Neural Networks
• A Tutorial on Deep Latent Variable Models of Natural Language
• Understanding the Limitations of Variational Mutual Information Estimators
• Robust Out-of-distribution Detection for Neural Networks
• MGGAN: Solving Mode Collapse using Manifold Guided Training
• Uncertainty-aware performance assessment of optical imaging modalities with invertible neural networks
• VideoFlow: A Conditional Flow-Based Model for Stochastic Video Generation
• Cross-Camera Feature Prediction for Intra-Camera Supervised Person Re-identification across Distant Scenes
• FitVid: Overfitting in Pixel-Level Video Prediction
• A Flow-Based Neural Network for Time Domain Speech Enhancement
• Improved Training of Generative Adversarial Networks Using Representative Features
• FloWaveNet : A Generative Flow for Raw Audio
• Neural Canonical Transformation with Symplectic Flows
• Learning Multimodal Transition Dynamics for Model-Based Reinforcement Learning
• Likelihood-free MCMC with Amortized Approximate Ratio Estimators
• Hamiltonian Generative Networks
• Neural Manifold Ordinary Differential Equations
• Benchmarking Invertible Architectures on Inverse Problems
• DeepFlow: History Matching in the Space of Deep Generative Models
• Hierarchical Autoregressive Image Models with Auxiliary Decoders
• Perfect density models cannot guarantee anomaly detection
• Diffusion Probabilistic Models for 3D Point Cloud Generation
• Generalizing Hamiltonian Monte Carlo with Neural Networks
• Poincaré Wasserstein Autoencoder
• StoryGAN: A Sequential Conditional GAN for Story Visualization
• Learning the temporal evolution of multivariate densities via normalizing flows
• Contrastive Embedding for Generalized Zero-Shot Learning
• Permutation Invariant Graph Generation via Score-Based Generative Modeling
• Unsupervised in-distribution anomaly detection of new physics through conditional density estimation
• Self-Supervised Variational Auto-Encoders
• Multi-band MelGAN: Faster Waveform Generation for High-Quality Text-to-Speech
• Effective LHC measurements with matrix elements and machine learning
• TF-Replicator: Distributed Machine Learning for Researchers
• Semi-Conditional Normalizing Flows for Semi-Supervised Learning
• Benchmarking deep inverse models over time, and the neural-adjoint method
• A Spectral Energy Distance for Parallel Speech Synthesis
• Boosting Trust Region Policy Optimization by Normalizing Flows Policy
• Preconditioned training of normalizing flows for variational inference in inverse problems
• Learning Symmetries of Classical Integrable Systems
• Wasserstein Dependency Measure for Representation Learning
• Towards Nonlinear Disentanglement in Natural Data with Temporal Sparse Coding
• Learnable Explicit Density for Continuous Latent Space and Variational Inference
• White Paper Machine Learning in Certified Systems
• Temporal Non-Volume Preserving Approach to Facial Age-Progression and Age-Invariant Face Recognition
• Likelihood-free inference with an improved cross-entropy estimator
• Transformation Autoregressive Networks
• Generative Latent Flow
• VCNet and Functional Targeted Regularization For Learning Causal Effects of Continuous Treatments
• STEER: Simple Temporal Regularization For Neural ODEs
• Variational Inference with NoFAS: Normalizing Flow with Adaptive Surrogate for Computationally Expensive Models
• Stochastic Normalizing Flows
• Metropolis-Hastings Generative Adversarial Networks
• Deep Probabilistic Imaging: Uncertainty Quantification and Multi-modal Solution Characterization for Computational Imaging
• Euclideanizing Flows: Diffeomorphic Reduction for Learning Stable Dynamical Systems
• Generation of ice states through deep reinforcement learning
• Unsupervised Out-of-Distribution Detection with Batch Normalization
• Faster Uncertainty Quantification for Inverse Problems with Conditional Normalizing Flows
• Shaping Belief States with Generative Environment Models for RL
• Machine Learning Statistical Gravity from Multi-Region Entanglement Entropy
• A Generative Model of Galactic Dust Emission Using Variational Inference
• Stein Neural Sampler
• Infinitely Deep Bayesian Neural Networks with Stochastic Differential Equations
• Introduction to Normalizing Flows for Lattice Field Theory
• Visual Anomaly Detection for Images: A Survey
• Improving Exploration in Soft-Actor-Critic with Normalizing Flows Policies
• Learning Overlapping Representations for the Estimation of Individualized Treatment Effects
• ShapeFlow: Learnable Deformations Among 3D Shapes
• Randomized Value Functions via Multiplicative Normalizing Flows
• PixelVAE++: Improved PixelVAE with Discrete Prior
• Using a thousand optimization tasks to learn hyperparameter search strategies
• MAE: Mutual Posterior-Divergence Regularization for Variational AutoEncoders
• Training Invertible Neural Networks as Autoencoders
• Split-Brain Autoencoders: Unsupervised Learning by Cross-Channel Prediction
• Deep Generative Quantile-Copula Models for Probabilistic Forecasting
• Visual Relationship Detection using Scene Graphs: A Survey
• TraDE: Transformers for Density Estimation
• Neural Entropic Estimation: A faster path to mutual information estimation
• Inference via low-dimensional couplings
• Bayesian Image Reconstruction using Deep Generative Models
• Hyperbolic Deep Neural Networks: A Survey
• MALI: A memory efficient and reverse accurate integrator for Neural ODEs
• Training Deep Energy-Based Models with f-Divergence Minimization
• Invertible Denoising Network: A Light Solution for Real Noise Removal
• Multi-Target Emotional Voice Conversion With Neural Vocoders
• Bilateral Denoising Diffusion Models
• Gradients as Features for Deep Representation Learning
• Reweighted Expectation Maximization
• Globally Injective ReLU Networks
• Fully probabilistic quasar continua predictions near Lyman-α with conditional neural spline flows
• MintNet: Building Invertible Neural Networks with Masked Convolutions
• Drop, Swap, and Generate: A Self-Supervised Approach for Generating Neural Activity
• Leveraging exploration in off-policy algorithms via normalizing flows
• FlowSeq: Non-Autoregressive Conditional Sequence Generation with Generative Flow
• Autoregressive Models: What Are They Good For?
• How to Hallucinate Functional Proteins
• Convolutional Normalizing Flows
• Invertible DenseNets with Concatenated LipSwish
• Duplex Sequence-to-Sequence Learning for Reversible Machine Translation
• Temporal Normalizing Flows
• Powering Hidden Markov Model by Neural Network based Generative Models
• Understanding the (un)interpretability of natural image distributions using generative models
• Leveraging the Exact Likelihood of Deep Latent Variable Models
• Copula & Marginal Flows: Disentangling the Marginal from its Joint
• Exchangeable Neural ODE for Set Modeling
• Automatic structured variational inference
• Non-parametric estimation of Jensen-Shannon Divergence in Generative Adversarial Network training
• Coupling the reduced-order model and the generative model for an importance sampling estimator
• Discriminator Contrastive Divergence: Semi-Amortized Generative Modeling by Exploring Energy of the Discriminator
• Regularized Autoencoders via Relaxed Injective Probability Flow
• Flow-based Kernel Prior with Application to Blind Super-Resolution
• Transforming Gaussian Processes With Normalizing Flows
• ChartPointFlow for Topology-Aware 3D Point Cloud Generation
• Deep Normalization for Speaker Vectors
• The Neural Moving Average Model for Scalable Variational Inference of State Space Models
• Flow Models for Arbitrary Conditional Likelihoods
• Metropolis-Hastings view on variational inference and adversarial training
• Hybrid Models for Open Set Recognition
• Artificial Intelligence in Drug Discovery: Applications and Techniques
• Bottleneck Conditional Density Estimation
• Normalizing flows for random fields in cosmology
• Sum-Product-Quotient Networks
• Multimodal Safety-Critical Scenarios Generation for Decision-Making Algorithms Evaluation
• Neural Density Estimation and Likelihood-free Inference
• OneFlow: One-class flow for anomaly detection based on a minimal volume region
• Supervised Anomaly Detection based on Deep Autoregressive Density Estimators
• Memory-Efficient Network for Large-scale Video Compressive Sensing
• On Scalable and Efficient Computation of Large Scale Optimal Transport
• DUAL-GLOW: Conditional Flow-Based Generative Model for Modality Transfer
• Maximum Entropy Flow Networks
• VAEBM: A Symbiosis between Variational Autoencoders and Energy-based Models
• Semi-Supervised Learning with Normalizing Flows
• SustainBench: Benchmarks for Monitoring the Sustainable Development Goals with Machine Learning
• WaveNODE: A Continuous Normalizing Flow for Speech Synthesis
• Universal Value Density Estimation for Imitation Learning and Goal-Conditioned Reinforcement Learning
• Super-resolution Variational Auto-Encoders
• CMTS: Conditional Multiple Trajectory Synthesizer for Generating Safety-critical Driving Scenarios
• Hierarchical Autoregressive Modeling for Neural Video Compression
• PixelCNN Models with Auxiliary Variables for Natural Image Modeling
• Video Interpolation and Prediction with Unsupervised Landmarks
• Structured Output Learning with Conditional Generative Flows
• Convolutional Bipartite Attractor Networks
• Path Integral Sampler: a stochastic control approach for sampling
• Stochastic Neural Network with Kronecker Flow
• Set Flow: A Permutation Invariant Normalizing Flow
• Sensing Anomalies as Potential Hazards: Datasets and Benchmarks
• Information Theory in Density Destructors
• PortaSpeech: Portable and High-Quality Generative Text-to-Speech
• On the Need for Topology-Aware Generative Models for Manifold-Based Defenses
• Modeling the Gaia Color-Magnitude Diagram with Bayesian Neural Flows to Constrain Distance Estimates
• Implicit Policy for Reinforcement Learning
• Generative models for sampling of lattice field theories
• Self-Supervised Learning of Generative Spin-Glasses with Normalizing Flows
• Deep Variational Inference Without Pixel-Wise Reconstruction
• Multimodal Generative Models for Compositional Representation Learning
• Representational aspects of depth and conditioning in normalizing flows
• Trumpets: Injective Flows for Inference and Inverse Problems
• CDCGen: Cross-Domain Conditional Generation via Normalizing Flows and Adversarial Training
• Sequential Neural Posterior and Likelihood Approximation
• Variational Inference with Continuously-Indexed Normalizing Flows
• Prior Image-Constrained Reconstruction using Style-Based Generative Models
• Neural Spatio-Temporal Point Processes
• MaCow: Masked Convolutional Generative Flow
• Implicit Distributional Reinforcement Learning
• Approximating exponential family models (not single distributions) with a two-network architecture
• Temperature-steerable flows
• BaCOUn: Bayesian Classifers with Out-of-Distribution Uncertainty
• Active World Model Learning with Progress Curiosity
• Moser Flow: Divergence-based Generative Modeling on Manifolds
• Wasserstein variational gradient descent: From semi-discrete optimal transport to ensemble variational inference
• A Method to Model Conditional Distributions with Normalizing Flows
• Generative Hybrid Representations for Activity Forecasting with No-Regret Learning
• Improving Sequential Latent Variable Models with Autoregressive Flows
• How to GAN Event Unweighting
• Non-Gaussian Gaussian Processes for Few-Shot Regression
• Beyond Domain Adaptation: Unseen Domain Encapsulation via Universal Non-volume Preserving Models
• Max-Affine Spline Insights into Deep Generative Networks
• MetFlow: A New Efficient Method for Bridging the Gap between Markov Chain Monte Carlo and Variational Inference
• Approaching the Limit of Image Rescaling via Flow Guidance
• Modelling Bounded Rationality in Multi-Agent Interactions by Generalized Recursive Reasoning
• Pseudo-Rehearsal for Continual Learning with Normalizing Flows
• DEMI: Discriminative Estimator of Mutual Information
• Log-Likelihood Ratio Minimizing Flows: Towards Robust and Quantifiable Neural Distribution Alignment
• Deep Gaussian Markov Random Fields
• Probabilistic Software Modeling: A Data-driven Paradigm for Software Analysis
• Relative gradient optimization of the Jacobian term in unsupervised deep learning
• Improved Autoregressive Modeling with Distribution Smoothing
• Generalized Energy Based Models
• Surfing: Iterative optimization over incrementally trained deep networks
• Improving Evolutionary Strategies with Generative Neural Networks
• Scaling Up Machine Learning For Quantum Field Theory with Equivariant Continuous Flows
• Sparse Flows: Pruning Continuous-depth Models
• A Probability Density Theory for Spin-Glass Systems
• Unsupervised Learning of Syntactic Structure with Invertible Neural Projections
• Towards Principled Uncertainty Estimation for Deep Neural Networks
• Deep Generative Learning via Schrödinger Bridge
• Quinoa: a Q-function You Infer Normalized Over Actions
• FINO: Flow-based Joint Image and Noise Model
• Are Deep Neural Architectures Losing Information? Invertibility Is Indispensable
• Generalized Zero-Shot Learning via VAE-Conditioned Generative Flow
• A Neural Network MCMC sampler that maximizes Proposal Entropy
• Trust-Region Variational Inference with Gaussian Mixture Models
• Conditional Normalizing Flows for Low-Dose Computed Tomography Image Reconstruction
• How well does your sampler really work?
• Flow Contrastive Estimation of Energy-Based Models
• Nonreversible MCMC from conditional invertible transforms: a complete recipe with convergence guarantees
• Secure Data Sharing With Flow Model
• An In-depth Summary of Recent Artificial Intelligence Applications in Drug Design
• Robust normalizing flows using Bernstein-type polynomials
• Variational Inference with Holder Bounds
• A novel stellar spectrum denoising method based on deep Bayesian modeling
• Automating Inference of Binary Microlensing Events with Neural Density Estimation
• The Autodidactic Universe
• Deep Generative Models Strike Back! Improving Understanding and Evaluation in Light of Unmet Expectations for OoD Data
• MeshODE: A Robust and Scalable Framework for Mesh Deformation
• Backpropagation for Implicit Spectral Densities
• Featurized Density Ratio Estimation
• Divergence Triangle for Joint Training of Generator Model, Energy-based Model, and Inference Model
• Fine-grained Sentiment Controlled Text Generation
• Resembled Generative Adversarial Networks: Two Domains with Similar Attributes
• Classify and Generate: Using Classification Latent Space Representations for Image Generations
• Learning nonlinear level sets for dimensionality reduction in function approximation
• A prior-based approximate latent Riemannian metric
• Problems using deep generative models for probabilistic audio source separation
• Video Extrapolation with an Invertible Linear Embedding
• Identifying through Flows for Recovering Latent Representations
• VAE-KRnet and its applications to variational Bayes
• Task-agnostic Continual Learning with Hybrid Probabilistic Models
• FaceShapeGene: A Disentangled Shape Representation for Flexible Face Image Editing
• Filtering Variational Objectives
• Semi-supervised Learning by Latent Space Energy-Based Model of Symbol-Vector Coupling
• Improving the Reconstruction of Disentangled Representation Learners via Multi-Stage Modeling
• MPG: A Multi-ingredient Pizza Image Generator with Conditional StyleGANs
• Measure Transport with Kernel Stein Discrepancy
• Scaling-up Diverse Orthogonal Convolutional Networks with a Paraunitary Framework
• Inference and De-Noising of Non-Gaussian Particle Distribution Functions: A Generative Modeling Approach
• EditVAE: Unsupervised Part-Aware Controllable 3D Point Cloud Shape Generation
• MaGNET: Uniform Sampling from Deep Generative Network Manifolds Without Retraining
• Fair Normalizing Flows
• Learning Space Partitions for Path Planning
• Toward Interactive Modulation for Photo-Realistic Image Restoration
• Inverse Aerodynamic Design of Gas Turbine Blades using Probabilistic Machine Learning
• Generative Probabilistic Image Colorization
• Learning Non-linear Wavelet Transformation via Normalizing Flow
• Autoregressive Score Matching
• From Image to Imuge: Immunized Image Generation
• Diverse Plausible Shape Completions from Ambiguous Depth Images
• SentenceMIM: A Latent Variable Language Model
• Woodbury Transformations for Deep Generative Flows
• Graph-based Normalizing Flow for Human Motion Generation and Reconstruction
• FIS-GAN: GAN with Flow-based Importance Sampling
• ASR-Free Pronunciation Assessment
• Two equalities expressing the determinant of a matrix in terms of expectations over matrix-vector products
• Geometrically Enriched Latent Spaces
• Variational Mixture of Normalizing Flows
• Semi-Implicit Generative Model
• Density Matching for Bilingual Word Embedding
• General Probabilistic Surface Optimization and Log Density Estimation
• Energy-Inspired Models: Learning with Sampler-Induced Distributions
• MIM: Mutual Information Machine
• A Forest from the Trees: Generation through Neighborhoods
• UNO: Uncertainty-aware Noisy-Or Multimodal Fusion for Unanticipated Input Degradation
• Kernel of CycleGAN as a Principle homogeneous space
• Recurrent Convolution for Compact and Cost-Adjustable Neural Networks: An Empirical Study
• Variational Bayes: A report on approaches and applications
• DOI: Divergence-based Out-of-Distribution Indicators via Deep Generative Models
• Training Invertible Linear Layers through Rank-One Perturbations
• Copula-Based Normalizing Flows
• TzK Flow - Conditional Generative Model
• Random Sum-Product Forests with Residual Links
• Visual Attention in Imaginative Agents
• An Iterative Closest Points Approach to Neural Generative Models
• PassFlow: Guessing Passwords with Generative Flows
• A Tale of Three Probabilistic Families: Discriminative, Descriptive and Generative Models
• Deep Residual Flow for Out of Distribution Detection
• Symmetric Wasserstein Autoencoders
• Quasi-Autoregressive Residual (QuAR) Flows
• EMIXER: End-to-end Multimodal X-ray Generation via Self-supervision
• Hierarchical Video Generation for Complex Data
• Causal Autoregressive Flows
• Deep generative LDA
• Model Repair: Robust Recovery of Over-Parameterized Statistical Models
• Copula-like Variational Inference
• Motion Mappings for Continuous Bilateral Teleoperation
• A Compact Network Learning Model for Distribution Regression
• Recurrent Estimation of Distributions
• Machine learning approaches for analyzing and enhancing molecular dynamics simulations
• Neural Discriminant Analysis for Deep Speaker Embedding
• Deep Image Synthesis from Intuitive User Input: A Review and Perspectives
• Discrete Point Flow Networks for Efficient Point Cloud Generation
• End-to-End Image Compression with Probabilistic Decoding
• D2C: Diffusion-Denoising Models for Few-shot Conditional Generation
• Probabilistic Modeling for Novelty Detection with Applications to Fraud Identification
• Improving Lossless Compression Rates via Monte Carlo Bits-Back Coding
• Re-examination of the Role of Latent Variables in Sequence Modeling
• Decoupling Global and Local Representations via Invertible Generative Flows
• Black-Box Inference for Non-Linear Latent Force Models
• On Investigation of Unsupervised Speech Factorization Based on Normalization Flow
• Convex Smoothed Autoencoder-Optimal Transport model
• Arbitrary Conditional Distributions with Energy
• General Invertible Transformations for Flow-based Generative Modeling
• Differentiate Everything with a Reversible Embeded Domain-Specific Language
• Normalizing Flows as a Novel PDF Turbulence Model
• Variational Determinant Estimation with Spherical Normalizing Flows
• Neural density estimation and uncertainty quantification for laser induced breakdown spectroscopy spectra
• Exploring Adversarial Examples via Invertible Neural Networks
• A Temporal Kernel Approach for Deep Learning with Continuous-time Information
• Universal Joint Approximation of Manifolds and Densities by Simple Injective Flows
• Inverse Learning of Symmetries
• Noise Conditional Flow Model for Learning the Super-Resolution Space
• Augmented KRnet for density estimation and approximation
• Flow-based Generative Models for Learning Manifold to Manifold Mappings
• Hierarchical Reinforcement Learning with Optimal Level Synchronization Based on Flow-Based Deep Generative Model
• NP-DRAW: A Non-Parametric Structured Latent Variable Model for Image Generation
• TzK: Flow-Based Conditional Generative Model
• Generating the support with extreme value losses
• Invertible Surrogate Models: Joint surrogate modelling and reconstruction of Laser-Wakefield Acceleration by invertible neural networks
• Bundle Networks: Fiber Bundles, Local Trivializations, and a Generative Approach to Exploring Many-to-one Maps
• Invertible Attention
• EBMs Trained with Maximum Likelihood are Generator Models Trained with a Self-adverserial Loss
• Learning to Shift Attention for Motion Generation
• Perceptual Image Restoration with High-Quality Priori and Degradation Learning
• Convex Nonparanormal Regression
• Integral Equations and Machine Learning
• Out-of-Distribution Detection of Melanoma using Normalizing Flows
• Flow-based Deformation Guidance for Unpaired Multi-Contrast MRI Image-to-Image Translation
• Gradient Boosted Normalizing Flows
• Generative Adversarial Learning via Kernel Density Discrimination
• Testing for Typicality with Respect to an Ensemble of Learned Distributions
• Natural Image Manipulation for Autoregressive Models Using Fisher Scores
• KGAN: How to Break The Minimax Game in GAN
• Latent Transformations for Discrete-Data Normalising Flows
• Synthetic Data Generation for Economists
• Can We Trust Deep Speech Prior?
• Recurrent Flow Networks: A Recurrent Latent Variable Model for Density Modelling of Urban Mobility
• Learning to Learn to Compress
• Generative Model with Dynamic Linear Flow
• Deep generative factorization for speech signal
• Cascading Modular Network (CAM-Net) for Multimodal Image Synthesis
• Uncertainty quantification for ptychography using normalizing flows
• Entropy-based adaptive Hamiltonian Monte Carlo
• Neural Network based Explicit Mixture Models and Expectation-maximization based Learning
• Data-driven Regularization via Racecar Training for Generalizing Neural Networks
• Predicting Classification Accuracy When Adding New Unobserved Classes
• Double Forward Propagation for Memorized Batch Normalization
• Imitation with Neural Density Models
• Self Normalizing Flows
• Time Series Source Separation with Slow Flows
• Autoregressive flow-based causal discovery and inference
• Bootstrap Your Flow
• Prior Flow Variational Autoencoder: A density estimation model for Non-Intrusive Load Monitoring
• Improving Compositionality of Neural Networks by Decoding Representations to Inputs
• RBM-Flow and D-Flow: Invertible Flows with Discrete Energy Base Spaces
• Generating Data Augmentation samples for Semantic Segmentation of Salt Bodies in a Synthetic Seismic Image Dataset
• Stay Positive: Non-Negative Image Synthesis for Augmented Reality
• Diffusion Normalizing Flow
• Computer Vision and Normalizing Flow-Based Defect Detection
• BPLF: A Bi-Parallel Linear Flow Model for Facial Expression Generation from Emotion Set Images
• Flexible Prior Distributions for Deep Generative Models
• Label-Conditioned Next-Frame Video Generation with Neural Flows
• Lossless Image Compression Using a Multi-Scale Progressive Statistical Model
• iVPF: Numerical Invertible Volume Preserving Flow for Efficient Lossless Compression
• Meta-analysis of dichotomous and ordinal tests without a gold standard
• X2CT-FLOW: Maximum a posteriori reconstruction using a progressive flow-based deep generative model for ultra sparse-view computed tomography in ultra low-dose protocols
• PointLIE: Locally Invertible Embedding for Point Cloud Sampling and Recovery
• Semantic and Geometric Unfolding of StyleGAN Latent Space
• Generative Actor-Critic: An Off-policy Algorithm Using the Push-forward Model
• Conditional Invertible Neural Networks for Diverse Image-to-Image Translation
• Disentangling Noise from Images: A Flow-Based Image Denoising Neural Network
• ADAVI: Automatic Dual Amortized Variational Inference Applied To Pyramidal Bayesian Models
• EMFlow: Data Imputation in Latent Space via EM and Deep Flow Models
• Hierarchical Policies for Cluttered-Scene Grasping with Latent Plans
• Equivariant Manifold Flows
• Mitigating Generation Shifts for Generalized Zero-Shot Learning