paper-with-me

Papers

Closing the Dequantization Gap: PixelCNN as a Single-Layer Flow

2020-02-06 · NeurIPS 2020 12 · Didrik Nielsen, Ole Winther

Flow models have recently made great progress at modeling ordinal discrete data such as images and audio. Due to the continuous nature of flow models, dequantization is typically applied when using them for such discrete data, resulting in lower bound estimates of the likelihood. In this paper, we introduce subset flows, a class of flows that can tractably transform finite volumes and thus allow exact computation of likelihoods for discrete data. Based on subset flows, we identify ordinal discrete autoregressive models, including WaveNets, PixelCNNs and Transformers, as single-layer flows. We use the flow formulation to compare models trained and evaluated with either the exact likelihood or its dequantization lower bound. Finally, we study multilayer flows composed of PixelCNNs and non-autoregressive coupling layers and demonstrate state-of-the-art results on CIFAR-10 for flow models trained with dequantization.

📄 PDF Abstract BibTeX arXiv:2002.02547

Code (1)

didriknielsen/pixelcnn_flow 공식 구현 pytorch

Methods 이 논문이 사용한 방법론

Mixture of Logistic Distributions 설명 없음
Dilated Causal Convolution A Dilated Causal Convolution is a causal convolution where the filter is applied over an area larger than its length by…
WaveNet WaveNet is an audio generative model based on the PixelCNN architecture. In order to deal with long-range temporal dependencies…
PixelCNN A PixelCNN is a generative model that uses autoregressive connections to model images pixel by pixel, decomposing the joint image distribution as a product of conditionals.…

Similar Papers 제목 키워드 기반

Conditional Image Generation with PixelCNN Decoders

2016-06-16 · NeurIPS 2016 12 · Aaron van den Oord, Nal Kalchbrenner, Oriol Vinyals, Lasse Espeholt 외

This work explores conditional image generation with a new image density model based on the PixelCNN architecture. The model can be conditioned on any vector, including descriptive labels or tags, or latent embeddings cr…

Conditional Image GenerationDecoderDensity EstimationDescriptive+1

Audio Dequantization for High Fidelity Audio Generation in Flow-based Neural Vocoder

2020-08-16 · Hyun-Wook Yoon, Sang-Hoon Lee, Hyeong-Rae Noh, Seong-Whan Lee

In recent works, a flow-based neural vocoder has shown significant improvement in real-time speech generation task. The sequence of invertible flow operations allows the model to convert samples from simple distribution …

Audio DequantizationAudio GenerationImage Generation

Flow++: Improving Flow-Based Generative Models with Variational Dequantization and Architecture Design

2019-02-01 · ICLR 2019 5 · Jonathan Ho, Xi Chen, Aravind Srinivas, Yan Duan 외

Flow-based generative models are powerful exact likelihood models with efficient sampling and inference. Despite their computational efficiency, flow-based models generally have much worse density modeling performance co…

Computational EfficiencyDensity EstimationImage Generation

Multi-Scale Architectures Matter: On the Adversarial Robustness of Flow-based Lossless Compression

2022-08-26 · Yi-chong Xia, Bin Chen, Yan Feng, Tian-shuo Ge

As a probabilistic modeling technique, the flow-based model has demonstrated remarkable potential in the field of lossless compression \cite{idf,idf++,lbb,ivpf,iflow},. Compared with other deep generative models (eg. Aut…

Adversarial RobustnessDensity Estimation

PixelVAE: A Latent Variable Model for Natural Images

2016-11-15 · Ishaan Gulrajani, Kundan Kumar, Faruk Ahmed, Adrien Ali Taiga 외

Natural image modeling is a landmark challenge of unsupervised learning. Variational Autoencoders (VAEs) learn a useful latent representation and model global structure well but have difficulty capturing small details. P…

Decodermodel