Jämför metoder
Granska de valda metoderna sida vid sida; rader som skiljer sig är markerade.
| Självövervakad Gaussisk process× | Gaussisk process× | |
|---|---|---|
| Ämnesområde | Maskininlärning | Maskininlärning |
| Familj | Machine learning | Machine learning |
| Ursprungsår≠ | 2019–2021 | 2006 (book); roots in Kriging, 1951) |
| Upphovsperson≠ | Fortuin, V. et al.; broader self-supervised GP literature | Rasmussen, C. E. & Williams, C. K. I. |
| Typ≠ | Probabilistic model (self-supervised GP pretraining + kernel learning) | Probabilistic non-parametric model |
| Ursprungskälla≠ | Fortuin, V., Rätsch, G., & Mandt, S. (2020). GP-VAE: Deep probabilistic time series imputation using Gaussian process variational autoencoders. Proceedings of the 23rd International Conference on Artificial Intelligence and Statistics (AISTATS), PMLR 108, 1651–1661. link ↗ | Rasmussen, C. E., & Williams, C. K. I. (2006). Gaussian Processes for Machine Learning. MIT Press. ISBN: 978-0-262-18253-9 |
| Alias | SSL-GP, self-supervised GP, self-supervised GPR, self-supervised Gaussian process regression | GP, Gaussian Process Regression, GPR, Kriging |
| Närliggande≠ | 6 | 3 |
| Sammanfattning≠ | Self-supervised Gaussian Process (SSL-GP) combines the principled uncertainty quantification of Gaussian processes with self-supervised pretraining, learning expressive kernels or latent representations from unlabeled data before fitting a GP on a small labeled set. This makes the approach especially powerful in low-labeled-data regimes where a conventional GP would overfit or produce poorly calibrated uncertainty estimates. | A Gaussian Process (GP) is a non-parametric, fully probabilistic machine learning model that places a prior distribution directly over functions. Rather than predicting a single value, it returns a predictive mean and a calibrated uncertainty estimate at every test point, making it especially valuable for regression on small to medium datasets and for Bayesian optimization tasks. |
| ScholarGateDatamängd ↗ |
|
|