pith. sign in

arxiv: 1609.02200 · v2 · pith:VXJR6KZ5new · submitted 2016-09-07 · 📊 stat.ML · cs.LG

Discrete Variational Autoencoders

classification 📊 stat.ML cs.LG
keywords discreteclasscomponentmodelsvariableslatentprobabilisticbackpropagation
0
0 comments X
read the original abstract

Probabilistic models with discrete latent variables naturally capture datasets composed of discrete classes. However, they are difficult to train efficiently, since backpropagation through discrete variables is generally not possible. We present a novel method to train a class of probabilistic models with discrete latent variables using the variational autoencoder framework, including backpropagation through the discrete latent variables. The associated class of probabilistic models comprises an undirected discrete component and a directed hierarchical continuous component. The discrete component captures the distribution over the disconnected smooth manifolds induced by the continuous component. As a result, this class of models efficiently learns both the class of objects in an image, and their specific realization in pixels, from unsupervised data, and outperforms state-of-the-art methods on the permutation-invariant MNIST, Omniglot, and Caltech-101 Silhouettes datasets.

This paper has not been read by Pith yet.

discussion (0)

Sign in with ORCID, Apple, or X to comment. Anyone can read and Pith papers without signing in.

Forward citations

Cited by 7 Pith papers

Reviewed papers in the Pith corpus that reference this work. Sorted by Pith novelty score.

  1. Policy Optimization in Hybrid Discrete-Continuous Action Spaces via Mixed Gradients

    cs.LG 2026-05 unverdicted novelty 7.0

    HPO enables unbiased policy optimization in hybrid action spaces by mixing differentiable simulation gradients with score-function estimates, outperforming PPO as continuous dimensions increase.

  2. Multi-Mode Quantum Annealing for Generative Representation Learning with Boltzmann Priors

    quant-ph 2026-04 unverdicted novelty 7.0

    A multi-mode quantum annealing approach enables VAEs with Boltzmann priors, showing faster training and better generation than Gaussian-prior VAEs on MNIST, Fashion-MNIST, and CelebA plus improved out-of-distribution ...

  3. DSA: Dynamic Step Allocation for Fast Autoregressive Video Generation

    cs.CV 2026-06 unverdicted novelty 6.0

    DSA adds a jointly trained confidence head to autoregressive video diffusion models that dynamically allocates fewer or more denoising steps per frame, achieving 22.63 FPS real-time generation on H100 while matching V...

  4. Scaling Autoregressive Models for Content-Rich Text-to-Image Generation

    cs.CV 2022-06 unverdicted novelty 6.0

    Scaling an autoregressive Transformer to 20B parameters for text-to-image generation using image token sequences achieves new SOTA zero-shot FID of 7.23 and fine-tuned FID of 3.22 on MS-COCO.

  5. Molecular Design beyond Training Data with Novel Extended Objective Functionals of Generative AI Models Driven by Quantum Annealing Computer

    q-bio.QM 2026-02 unverdicted novelty 5.0

    Quantum annealing combined with a Neural Hash Function lets generative models create molecules that are more drug-like than classical versions or the training set itself.

  6. A Survey on Deep Learning Architectures for Point Cloud Classification and Segmentation

    cs.CV 2026-05 unverdicted novelty 4.0

    A systematic literature survey that categorizes deep learning architectures for point cloud classification, part segmentation, and semantic segmentation, evaluates them on benchmarks, and discusses innovations, limita...

  7. A Survey on Deep Learning Architectures for Point Cloud Classification and Segmentation

    cs.CV 2026-05 unverdicted novelty 1.0

    A survey that categorizes deep learning models for point cloud tasks by backbone architecture, evaluates benchmark performance, and outlines challenges and future research directions.