A locally-built, LLM-digested index of recent arXiv papers in quant finance, geometry/topology, and statistical ML — keyword search served straight from SQLite on this machine.
tvGP-VAE models tensor-valued latent variables with Gaussian processes for better data structure representation.
The paper provides convergence guarantees for VAEs using SGD and Adam.
Integrates VAEs into EM for deep clustering and generation.
PriorVAE uses VAEs to efficiently encode spatial priors for small-area estimation.
This paper proposes a method to improve VAEs by extracting latent spaces from pre-trained diffusion models.
Develops SGP-VAE for efficient sparse GP inference in multi-dimensional datasets.
Proposes a VAE variant for ordinal content factors.
Deep generative models are commonly used for generating images and text. Interpretability of these models is one important pursuit, other than the generation quality. Variational auto-encoder (VAE) with Gaussian distribution as prior has been successfully applied in text generation, but it is hard to interpret the mean…
We propose a family of novel hierarchical Bayesian deep auto-encoder models capable of identifying disentangled factors of variability in data. While many recent attempts at factor disentanglement have focused on sophisticated learning objectives within the VAE framework, their choice of a standard normal as the latent…
We study a variant of the variational autoencoder model (VAE) with a Gaussian mixture as a prior distribution, with the goal of performing unsupervised clustering through deep generative models. We observe that the known problem of over-regularisation that has been shown to arise in regular VAEs also manifests itself i…
Proposes a new prior for VAEs to improve out-of-distribution detection.
The variational autoencoder (VAE) is a generative model with continuous latent variables where a pair of probabilistic encoder (bottom-up) and decoder (top-down) is jointly learned by stochastic gradient variational Bayes. We first elaborate Gaussian VAE, approximating the local covariance matrix of the decoder as an o…
VAEs and GANs use simple distributions and neural networks to implicitly approximate complex data distributions.
ELBO of VAEs converges to a sum of three entropies.
Variational auto-encoders (VAEs) are a popular and powerful deep generative model. Previous works on VAEs have assumed a factorized likelihood model, whereby the output uncertainty of each pixel is assumed to be independent. This approximation is clearly limited as demonstrated by observing a residual image from a VAE …
Stochastic processes provide a mathematically elegant way model complex data. In theory, they provide flexible priors over function classes that can encode a wide range of interesting assumptions. In practice, however, efficient inference by optimisation or marginalisation is difficult, a problem further exacerbated wi…
Posterior collapse in Variational Autoencoders (VAEs) arises when the variational posterior distribution closely matches the prior for a subset of latent variables. This paper presents a simple and intuitive explanation for posterior collapse through the analysis of linear VAEs and their direct correspondence with Prob…
Although variational autoencoders (VAEs) represent a widely influential deep generative model, many aspects of the underlying energy function remain poorly understood. In particular, it is commonly believed that Gaussian encoder/decoder assumptions reduce the effectiveness of VAEs in generating realistic samples. In th…
Deep latent variable models (LVM) such as variational auto-encoder (VAE) have recently played an important role in text generation. One key factor is the exploitation of smooth latent structures to guide the generation. However, the representation power of VAEs is limited due to two reasons: (1) the Gaussian assumption…
HH-VAEM improves imputation and acquisition of missing data using hierarchical models and Hamiltonian Monte Carlo.
Our work improves VAE latent space clustering by enforcing invariant and equivariant learning.
A new approach to VAEs tackles variance shrinkage using quantile regression.
Using powerful posterior distributions is a popular approach to achieving better variational inference. However, recent works showed that the aggregated posterior may fail to match unit Gaussian prior, thus learning the prior becomes an alternative way to improve the lower-bound. In this paper, for the first time in th…
PH-VAE models heavy-tailed data with flexible Phase-Type distributions.
GMVAEs cluster and generate game levels without labels.
VAEs analyzed using harmonic analysis, showing how variance controls frequency content and robustness.
Proposes a method to learn conditional VAEs from datasets with missing covariates.
BS-VAE separates decoder variance and beta to improve VAE performance.
To address the challenges in learning deep generative models (e.g.,the blurriness of variational auto-encoder and the instability of training generative adversarial networks, we propose a novel deep generative model, named Wasserstein-Wasserstein auto-encoders (WWAE). We formulate WWAE as minimization of the penalized …
Proposes deep quantile regression for uncertainty estimation in lesion detection.
Flexible VAEs using FIFs improve model likelihood on image datasets.
Paper uses VAEs to detect radar targets in complex noise.
C-VAE improves class representation in long-tailed generative models.
Improved deep hierarchical VAE with diffusion-based VampPrior.
AEVB improves understanding of latent variable models.
CVAE learns disentangled and coupled representations without prior knowledge.
Multivariate time series with missing values are common in areas such as healthcare and finance, and have grown in number and complexity over the years. This raises the question whether deep learning methodologies can outperform classical data imputation methods in this domain. However, naive applications of deep learn…
Variational Autoencoders (VAEs) provide a theoretically-backed and popular framework for deep generative models. However, learning a VAE from data poses still unanswered theoretical questions and considerable practical challenges. In this work, we propose an alternative framework for generative modeling that is simpler…
X-VAE uses data-adaptive Gaussian priors to improve latent space modeling.
Class labels are often imperfectly observed, due to mistakes and to genuine ambiguity among classes. We propose a new semi-supervised deep generative model that explicitly models noisy labels, called the Mislabeled VAE (M-VAE). The M-VAE can perform better than existing deep generative models which do not account for l…
New analysis enables inversion of deep generative models with unique solutions.
Generative Adversarial Nets (GANs) and Variational Auto-Encoders (VAEs) provide impressive image generations from Gaussian white noise, but the underlying mathematics are not well understood. We compute deep convolutional network generators by inverting a fixed embedding operator. Therefore, they do not require to be o…
New GP-VAE model improves scalability and performance.
PDGMM-VAE uses adaptive priors for better ICA recovery.
In narrow asymptotic settings Gaussian VAE models of continuous data have been shown to possess global optima aligned with ground-truth distributions. Even so, it is well known that poor solutions whereby the latent posterior collapses to an uninformative prior are sometimes obtained in practice. However, contrary to c…
Conventional prior for Variational Auto-Encoder (VAE) is a Gaussian distribution. Recent works demonstrated that choice of prior distribution affects learning capacity of VAE models. We propose a general technique (embedding-reparameterization procedure, or ER) for introducing arbitrary manifold-valued variables in VAE…
A tutorial on VAEs explaining its derivation and applications.
Proposes L-VAE for longitudinal data analysis.