Generative Modeling on MNIST real-valued (test)
3,402Log-likelihoodAdversarial Autoencoder
Evaluation Results
| Method | Links | |
|---|---|---|
| Adversarial AutoencoderParzen window samples=10K2015.11 | 3,402 | |
| GMMN + AEParzen window samples=10K2015.11 | 2,822 | |
| GANParzen window samples=10K2015.11 | 2,252 | |
| Deep GSNParzen window samples=10K2015.11 | 2,141.1 | |
| DBNParzen window samples=10K2015.11 | 1,382 | |
| Stacked CAEParzen window samples=10K2015.11 | 1,211.6 | |
| Adversarial AutoencoderParzen window samples=10M2015.11 | 427 | |
| GANParzen window samples=10M2015.11 | 386 |