Tractable Density Estimation on Learned Manifolds with Conformal Embedding Flows
arXiv:2106.05275
Abstract
Normalizing flows are generative models that provide tractable density estimation via an invertible transformation from a simple base distribution to a complex target distribution. However, this technique cannot directly model data supported on an unknown low-dimensional manifold, a common occurrence in real-world domains such as image data. Recent attempts to remedy this limitation have introduced geometric complications that defeat a central benefit of normalizing flows: exact density estimation. We recover this benefit with Conformal Embedding Flows, a framework for designing flows that learn manifolds with tractable densities. We argue that composing a standard flow with a trainable conformal embedding is the most natural way to model manifold-supported data. To this end, we present a series of conformal building blocks and apply them in experiments with synthetic and real-world data to demonstrate that flows can model manifold-supported distributions without sacrificing tractable likelihoods.
NeurIPS 2021 Camera-Ready. Code: https://github.com/layer6ai-labs/CEF
References in corpus (12)
- GANs Trained by a Two Time-Scale Update Rule Converge to a Local Nash Equilibrium
- NICE: Non-linear Independent Components Estimation
- Orthogonal Weight Normalization: Solution to Optimization over Multiple Dependent Stiefel Manifolds in Deep Neural Networks
- Improving Variational Auto-Encoders using Householder Flow
- Generalized BackPropagation, Étude De Cas: Orthogonality
- Optimization on Submanifolds of Convolution Kernels in CNNs
- Trivializations for Gradient-Based Optimization on Manifolds
- Deep Isometric Learning for Visual Recognition
- Regularized Autoencoders via Relaxed Injective Probability Flow
- Rectangular Flows for Manifold Learning
- Fair Normalizing Flows
- Generalization of the Change of Variables Formula with Applications to Residual Flows