Abstrakti
We propose non-stationary spectral kernels for Gaussian process regression. We propose to model the spectral density of a non-stationary kernel function as a mixture of input-dependent Gaussian process frequency density surfaces. We solve the generalised Fourier transform with such a model, and present a family of non-stationary and non-monotonic kernels that can learn input-dependent and potentially long-range, non-monotonic covariances between inputs. We derive efficient inference using model whitening and marginalized posterior, and show with case studies that these kernels are necessary when modelling even rather simple time series, image or geospatial data with non-stationary characteristics.
Alkuperäiskieli | Englanti |
---|---|
Otsikko | Advances in Neural Information Processing Systems 30 |
Alaotsikko | Proceedings of NIPS2017 |
Kustantaja | Curran Associates, Inc. |
Sivut | 4645-4654 |
Tila | Julkaistu - 2017 |
OKM-julkaisutyyppi | A4 Artikkeli konferenssijulkaisuussa |
Tapahtuma | NIPS Symposium on Interpretable Machine Learning - Long Beach, Los Angeles, Yhdysvallat Kesto: 4 joulukuuta 2017 → 9 joulukuuta 2017 Konferenssinumero: 31 |
Julkaisusarja
Nimi | Advances in Neural Information Processing Systems |
---|---|
Kustantaja | Curran Associates |
Vuosikerta | 30 |
ISSN (painettu) | 1049-5258 |
Conference
Conference | NIPS Symposium on Interpretable Machine Learning |
---|---|
Maa | Yhdysvallat |
Kaupunki | Los Angeles |
Ajanjakso | 04/12/2017 → 09/12/2017 |