Neural Autoregressive Flows
This work addresses the need for more expressive probabilistic models in machine learning, offering an incremental improvement over prior autoregressive flow techniques.
The paper tackled the problem of improving density estimation and variational autoencoders by unifying and generalizing existing autoregressive flow methods, resulting in state-of-the-art performance on density estimation tasks and outperforming Inverse Autoregressive Flows on binarized MNIST.
Normalizing flows and autoregressive models have been successfully combined to produce state-of-the-art results in density estimation, via Masked Autoregressive Flows (MAF), and to accelerate state-of-the-art WaveNet-based speech synthesis to 20x faster than real-time, via Inverse Autoregressive Flows (IAF). We unify and generalize these approaches, replacing the (conditionally) affine univariate transformations of MAF/IAF with a more general class of invertible univariate transformations expressed as monotonic neural networks. We demonstrate that the proposed neural autoregressive flows (NAF) are universal approximators for continuous probability distributions, and their greater expressivity allows them to better capture multimodal target distributions. Experimentally, NAF yields state-of-the-art performance on a suite of density estimation tasks and outperforms IAF in variational autoencoders trained on binarized MNIST.