1D Regression on Synthetic 1D Regression Periodic kernel GP
1.372Context LikelihoodNeural Processes with Stochastic Attention
Evaluation Results
| Method | Links | ||
|---|---|---|---|
| Neural Processes with Stochastic Attentionregularization=proposed2022.04 | 1.372 | -0.612 | |
| ConvCNPrepresentation=functional2022.04 | 1.358 | -0.502 | |
| ConvNPrepresentation=functional2022.04 | 0.953 | -0.522 | |
| ANPWeight decay lambda=0.0012022.04 | 0.926 | -0.644 | |
| ANPregularization=bootstrapping2022.04 | 0.711 | -0.717 | |
| ANPImportance Weighted ELBO samples=52022.04 | 0.704 | -0.687 | |
| CNPaggregation=Bayesian2022.04 | 0.278 | -1.026 | |
| CNPImportance Weighted ELBO samples=52022.04 | 0.231 | -0.957 | |
| CNPWeight decay lambda=0.0012022.04 | 0.185 | -0.97 | |
| CNP2022.04 | 0.176 | -0.978 | |
| ANP2022.04 | 0.111 | -0.951 | |
| ANPregularization=dropout2022.04 | 0.014 | -0.877 | |
| NPImportance Weighted ELBO samples=52022.04 | -0.33 | -1.094 | |
| NP2022.04 | -0.4 | -1.321 | |
| NPWeight decay lambda=0.0012022.04 | -0.401 | -1.315 | |
| NPaggregation=Bayesian2022.04 | -0.537 | -0.877 |