首页/论文/1908.09948

arXiv 1908.099482019-08-26cs.CVcs.LGstat.ML

PixelVAE++: Improved PixelVAE with Discrete Prior

D-Wave Systems Inc.（D-Wave系统公司）；University of Florence（佛罗伦萨大学）；Simon Fraser University（西蒙弗雷泽大学）

Hossein Sadeghi, Evgeny Andriyash, Walter Vinci, Lorenzo Buffoni, Mohammad H. Amin

英文摘要：

Constructing powerful generative models for natural images is a challenging task. PixelCNN models capture details and local information in images very well but have limited receptive field. Variational autoencoders with a factorial decoder can capture global information easily, but they often fail to reconstruct details faithfully. PixelVAE combines the best features of the two models and constructs a generative model that is able to learn local and global structures. Here we introduce PixelVAE++, a VAE with three types of latent variables and a PixelCNN++ for the decoder. We introduce a novel architecture that reuses a part of the decoder as an encoder. We achieve the state of the art performance on binary data sets such as MNIST and Omniglot and achieve the state of the art performance on CIFAR-10 among latent variable models while keeping the latent variables informative.

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

阅读原文

阅读 PDF ↗

arXiv 页面 ↗HTML 原文 ↗

本文内容

英文摘要相关论文

机构、总结、中文摘要均由AI分析而来，可能存在错误，请以论文正文为准。

继续探索

相关论文

为什么掩码扩散有效：以跳跃时间表为条件以改进离散扩散

Why Masking Diffusion Works: Condition on the Jump Schedule for Improved Discrete Diffusion

New York University（纽约大学）

本文揭示掩码扩散因内置已知跳跃时间分布而表现优异，并提出 SCUD 将该机制推广到任意离散扩散模型，在图像、文本和蛋白质数据上超越掩码扩散。

2025-09-27cs.LGstat.ML

利用纠错码改进离散 VAE 中的变分推断

Improved Variational Inference in Discrete VAEs using Error Correcting Codes

Dept. of Signal Theory and Communications, Universidad Rey Juan Carlos（信号理论与通信系，雷乌安卡洛斯大学）；Klipsch School of Electrical and Computer Engineering, New Mexico State University（电气与计算机工程学院，新墨西哥州立大学）；Instituto de Investigación Sanitaria Gregorio Marañón, Madrid, Spain（格里戈里奥·马兰农卫生研究所，马德里，西班牙）

该研究将离散 VAE 视为通信系统，利用 ECC 为离散潜在表示引入冗余，从而获得更准确的变分后验，并提升生成、重建与校准性能。

2025-06-10cs.LG

改进的离散扩散模型采样方案

Improved Sampling Schedules for Discrete Diffusion Models

Department of Data Science, EURECOM（数据科学系，EURECOM）；Donders Institute for Brain, Cognition and Behaviour（脑科学与行为研究所）；Radboud University（拉德堡德大学）

本文提出基于熵产率和Wasserstein距离的两种改进采样方案，提升离散扩散模型在多个领域的生成性能。

2026-02-09cs.LG

分裂吉布斯离散扩散后验采样

Split Gibbs Discrete Diffusion Posterior Sampling

California Institute of Technology（加州理工学院）；New York University（纽约大学）；OpenAI

提出基于分裂吉布斯采样的离散扩散后验采样算法SGDD，支持奖励引导生成与逆问题求解，在多项基准上取得SOTA性能及30%图像修复提升。

2025-11-01cs.CVcs.LG

贝叶斯神经网络深度估计的改进

Improved Depth Estimation of Bayesian Neural Networks

Lazy Dynamics（拉齐动力学）；Eindhoven University of Technology（埃因霍温理工大学）；GN Hearing（GN听力）

本文针对贝叶斯神经网络深度估计问题，提出基于离散截断正态分布独立学习网络深度均值与方差，并通过最小化变分自由能推断后验，有效提升了螺旋数据集测试精度并降低了深度估计方差。

2024-10-15cs.LGstat.ML

DGSAN: Discrete Generative Self-Adversarial Network

Sharif University of Technology（谢里夫理工大学）

Although GAN-based methods have received many achievements in the last few years, they have not been entirelysuccessful in generating discrete data. The most crucial challenge of these methods is the difficulty of passing the gradientfrom the discriminator to the generator when the generator outputs are discrete. Despite the fact that several attemptshave been made to alleviate this problem, none of the existing GAN-based methods have improved the performance oftext generation compared with the maximum likelihood approach in terms of both the quality and the diversity. In thispaper, we propose…

2020-10-15cs.LGstat.ML

更多相关论文收起更多论文

Oops I Took A Gradient: Scalable Sampling for Discrete Distributions

University of Toronto（多伦多大学）；Vector Institute（向量研究所）；Google Research（谷歌研究院）

We propose a general and scalable approximate sampling strategy for probabilistic models with discrete variables. Our approach uses gradients of the likelihood function with respect to its discrete inputs to propose updates in a Metropolis-Hastings sampler. We show empirically that this approach outperforms generic samplers in a number of difficult settings including Ising models, Potts models, restricted Boltzmann machines, and factorial hidden Markov models. We also demonstrate the use of our improved sampler for training deep energy-based models on high dimensional discrete data. This appro…

2021-06-06cs.LG

Scalable Gaussian Processes on Discrete Domains

University College London（伦敦大学学院）

Kernel methods on discrete domains have shown great promise for many challenging data types, for instance, biological sequence data and molecular structure data. Scalable kernel methods like Support Vector Machines may offer good predictive performances but do not intrinsically provide uncertainty estimates. In contrast, probabilistic kernel methods like Gaussian Processes offer uncertainty estimates in addition to good predictive performance but fall short in terms of scalability. While the scalability of Gaussian processes can be improved using sparse inducing point approximations, the selec…

2021-05-26cs.AIcs.LGstat.ML

离散状态空间上的生成流：实现多模态流及其在蛋白质协同设计中的应用

Generative Flows on Discrete State-Spaces: Enabling Multimodal Flows with Applications to Protein Co-Design

University of Oxford（牛津大学）；Massachusetts Institute of Technology（麻省理工学院）

本文提出离散流模型（DFMs），利用连续时间马尔可夫链实现离散数据的流匹配，统一了离散扩散模型，并构建多模态流框架，在蛋白质序列与结构协同设计任务中取得最先进性能。

2024-06-05cs.LGq-bio.QMstat.ML

使用离散 Walk-Jump 采样进行蛋白质发现

Protein Discovery with Discrete Walk-Jump Sampling

Prescient Design（普雷赛特设计公司）；Genentech（基因泰克公司）；New York University（纽约大学）

本文提出离散 Walk-Jump Sampling，用单噪声水平的平滑能量建模、Langevin MCMC 与一步去噪实现抗体蛋白质生成，并以分布一致性分数显著提升实验成功率和结合亲和力。

2024-03-15cs.LGq-bio.BM

确定性离散去噪

Deterministic Discrete Denoising

Graduate School of Information Science and Technology, The University of Osaka, Osaka, Japan（信息科学与技术研究生学校，大阪大学，大阪，日本）；International Research Center for Neurointelligence, The University of Tokyo, Tokyo, Japan（神经智能国际研究中心，东京大学，东京，日本）

本文提出了一种基于herding算法的确定性去噪方法，用于提升离散扩散模型的生成效率与质量。

2026-01-29cs.LGnlin.CD

基于分类器引导的germline吸收离散扩散的抗体序列生成

Conditional generation of antibody sequences with classifier-guided germline-absorbing discrete diffusion

Paul G. Allen School of Computer Science and Engineering, University of Washington（华盛顿大学保罗·G·艾伦计算机科学与工程学院）；Life Sciences, Amazon Web Services（亚马逊网络服务生命科学）

本文提出germline吸收扩散模型，通过离散扩散微调提升抗体序列生成性能，有效减少germline偏差，改进非germline残基预测准确率，并在条件生成任务中优于现有方法。

2026-05-07cs.AIcs.LG

Drax：基于离散流匹配的语音识别

Drax: Speech Recognition with Discrete Flow Matching

本研究针对非自回归模型在自动语音识别领域潜力未充分挖掘的问题，提出离散流匹配框架Drax，通过构建音频条件概率路径对齐训练与推理，实现了与SOTA相当的准确率及更优的准确率-效率权衡。

2025-10-05cs.LGcs.SDeess.AS

SJD++: 改进的推测雅可比解码用于无训练加速离散自回归文本到图像生成

SJD++: Improved Speculative Jacobi Decoding for Training-free Acceleration of Discrete Auto-regressive Text-to-Image Generation

The University of Hong Kong（香港大学）；Huawei Noah’s Ark Lab（华为诺亚实验室）；Peking University（北京大学）；xAI；Tsinghua University（清华大学）；Shanghai Jiao Tong University（上海交通大学）

SJD++通过概率并行解码算法提升自回归文本到图像生成的效率，实现2-3倍的推理延迟降低和2-7倍的步骤压缩，同时保持图像质量。

2025-12-08cs.CV

Parallel D2-Clustering: Large-Scale Clustering of Discrete Distributions

The Pennsylvania State University（宾夕法尼亚州立大学）

The discrete distribution clustering algorithm, namely D2-clustering, has demonstrated its usefulness in image classification and annotation where each object is represented by a bag of weighed vectors. The high computational complexity of the algorithm, however, limits its applications to large-scale problems. We present a parallel D2-clustering algorithm with substantially improved scalability. A hierarchical structure for parallel computing is devised to achieve a balance between the individual-node computation and the integration process of the algorithm. Additionally, it is shown that eve…

2013-02-06cs.CVcs.LG

Topology-Aware Segmentation Using Discrete Morse Theory

Stony Brook University（石溪大学）；University of California, San Diego（加利福尼亚大学圣迭戈分校）；Oregon State University（俄勒冈州立大学）

In the segmentation of fine-scale structures from natural and biomedical images, per-pixel accuracy is not the only metric of concern. Topological correctness, such as vessel connectivity and membrane closure, is crucial for downstream analysis tasks. In this paper, we propose a new approach to train deep image segmentation networks for better topological accuracy. In particular, leveraging the power of discrete Morse theory (DMT), we identify global structures, including 1D skeletons and 2D patches, which are important for topological accuracy. Trained with a novel loss based on these global …

2021-03-18cs.CGcs.CV