.png){kind=icon}
Return
Learning Generative Models Using Denoising Density Estimators
.png){kind=icon}
OA
AI
DOI：10.1109/TNNLS.2023.3308191.png)
.png){kind=icon}
Original request for help.png){kind=icon}
Share.png){kind=icon}
Save
Abstract
En 中文
Learning probabilistic models that can estimate the density of a given set of samples, and generate samples from that density, is one of the fundamental challenges in unsupervised machine learning. We introduce a new generative model based on denoising density estimators (DDEs), which are scalar functions parametrized by neural networks, that are efficiently trained to represent kernel density estimators of the data. Leveraging DDEs, our main contribution is a novel technique to obtain generative models by minimizing the Kullback-Leibler (KL)-divergence directly. We prove that our algorithm for obtaining generative models is guaranteed to converge consistently to the correct solution. Our approach does not require specific network architecture as in normalizing flows (NFs), nor use ordinary differential equation (ODE) solvers as in continuous NFs. Experimental results demonstrate substantial improvement in density estimation and competitive performance in generative model training.
Keywords:
Denoising autoencoders (DAEs)
density estimation
energy models
generative modeling
score-matching
Journal
IF:
8.9
Papers:
7.6K
Citations:
7.2W
Organization
No organization information available
Cited Papers
Deep associative neural network for associative memory based on unsupervised representation learning.png)
NEURAL NETWORKS
OAAI
NEURAL NETWORKSIF6.3
OAAIShareSaveShareSaveShareSaveShareSaveShareSaveShareSave
View more