Synthetic 1D Regression RBF kernel with noises
1.374Context LikelihoodNeural Processes with Stochastic Attention
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Neural Processes with Stochastic Attentionregularization=proposed2022.04 | 1.374 | -0.337 | |
| ConvCNPrepresentation=functional2022.04 | 1.314 | -0.428 | |
| ANPWeight decay lambda=0.0012022.04 | 0.957 | -0.442 | |
| ConvNPrepresentation=functional2022.04 | 0.873 | -0.469 | |
| ANPImportance Weighted ELBO samples=52022.04 | 0.771 | -0.47 | |
| ANPregularization=bootstrapping2022.04 | 0.754 | -0.407 | |
| CNPaggregation=Bayesian2022.04 | 0.351 | -0.508 | |
| CNPImportance Weighted ELBO samples=52022.04 | 0.271 | -0.442 | |
| CNPWeight decay lambda=0.0012022.04 | 0.24 | -0.471 | |
| CNP2022.04 | 0.233 | -0.478 | |
| ANP2022.04 | 0.228 | -0.603 | |
| ANPregularization=dropout2022.04 | 0.158 | -0.593 | |
| NP2022.04 | -0.151 | -0.69 | |
| NPWeight decay lambda=0.0012022.04 | -0.153 | -0.709 | |
| NPImportance Weighted ELBO samples=52022.04 | -0.155 | -0.633 | |
| NPaggregation=Bayesian2022.04 | -0.406 | -0.723 |