Density Estimation
The goal of Density Estimation is to give an accurate description of the underlying probabilistic density distribution of an observable data set with unknown density.
Source: Contrastive Predictive Coding Based Feature for Automatic Speaker Verification
Papers
Showing 1–10 of 1394 papers
All datasetsCIFAR-10MNISTUCI POWERBSDS300ImageNet 32x32UCI GASUCI HEPMASSUCI MINIBOONEOMNIGLOTCaltech-101FreyfacesCelebA-HQ 256x256
Benchmark Results
| # | Model | Metric | Claimed | Verified | Status |
|---|---|---|---|---|---|
| 1 | MAF | Log-likelihood (nats) | 3,049 | — | Unverified |
| 2 | DDPM | NLL (bits/dim) | 3.69 | — | Unverified |
| 3 | MRCNF | NLL (bits/dim) | 3.54 | — | Unverified |
| 4 | FFJORD | NLL (bits/dim) | 3.4 | — | Unverified |
| 5 | RNODE | NLL (bits/dim) | 3.38 | — | Unverified |
| 6 | Pixel CNN | NLL (bits/dim) | 3.03 | — | Unverified |
| 7 | score SDE | NLL (bits/dim) | 2.99 | — | Unverified |
| 8 | Flow matching | NLL (bits/dim) | 2.99 | — | Unverified |
| 9 | Pixel CNN ++ | NLL (bits/dim) | 2.92 | — | Unverified |
| 10 | Image Transformer | NLL (bits/dim) | 2.9 | — | Unverified |