Share your thoughts, 1 month free Claude Pro on usSee more
WorkDL logo mark

Auto-Encoding Variational Bayes

About

How can we perform efficient inference and learning in directed probabilistic models, in the presence of continuous latent variables with intractable posterior distributions, and large datasets? We introduce a stochastic variational inference and learning algorithm that scales to large datasets and, under some mild differentiability conditions, even works in the intractable case. Our contributions are two-fold. First, we show that a reparameterization of the variational lower bound yields a lower bound estimator that can be straightforwardly optimized using standard stochastic gradient methods. Second, we show that for i.i.d. datasets with continuous latent variables per datapoint, posterior inference can be made especially efficient by fitting an approximate inference model (also called a recognition model) to the intractable posterior using the proposed lower bound estimator. Theoretical advantages are reflected in experimental results.

Diederik P Kingma, Max Welling• 2013

Related benchmarks

TaskDatasetResultRank
Image ClassificationMNIST (test)
Accuracy54
894
Time Series ForecastingETTh1
MSE0.463
729
Anomaly LocalizationMVTec AD--
513
Image GenerationCIFAR-10 (test)
FID110.4
483
Image ClassificationMNIST
Accuracy94.5
417
Image ClassificationSVHN
Accuracy30.8
395
Time Series ForecastingETTh1 (test)
MSE0.681
348
Time Series ForecastingETTm1
MSE2.321
334
Image ClusteringCIFAR-10
NMI0.245
318
Image ClusteringSTL-10
ACC28.2
282
Showing 10 of 344 rows
...

Other info

Follow for update