1D Regression on Synthetic Matern Kernel GP
1.365Context LikelihoodNeural Processes with Stochastic Attention
Evaluation results
| Method | Links | ||
|---|---|---|---|
| Neural Processes with Stochastic Attentionregularization=proposed2022.04 | 1.365 | -0.175 | |
| ConvCNPrepresentation=functional2022.04 | 1.319 | -0.119 | |
| ANPWeight decay lambda=0.0012022.04 | 1.014 | -0.209 | |
| ConvNPrepresentation=functional2022.04 | 0.832 | -0.163 | |
| ANPImportance Weighted ELBO samples=52022.04 | 0.8 | -0.324 | |
| ANPregularization=bootstrapping2022.04 | 0.788 | -0.303 | |
| CNPaggregation=Bayesian2022.04 | 0.349 | -0.569 | |
| CNPImportance Weighted ELBO samples=52022.04 | 0.321 | -0.46 | |
| CNPWeight decay lambda=0.0012022.04 | 0.278 | -0.472 | |
| ANP2022.04 | 0.254 | -0.488 | |
| CNP2022.04 | 0.246 | -0.544 | |
| ANPregularization=dropout2022.04 | 0.136 | -0.575 | |
| NPWeight decay lambda=0.0012022.04 | -0.228 | -0.936 | |
| NP2022.04 | -0.235 | -0.918 | |
| NPImportance Weighted ELBO samples=52022.04 | -0.238 | -0.779 | |
| NPaggregation=Bayesian2022.04 | -0.45 | -0.837 |