Saved in:
| Main Authors: | , , , |
|---|---|
| Format: | Preprint |
| Published: |
2023
|
| Subjects: | |
| Online Access: | https://arxiv.org/abs/2311.08636 |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866916294699253760 |
|---|---|
| author | Kim, Keunsu Lyu, Hanbaek Kim, Jinsu Jung, Jae-Hun |
| author_facet | Kim, Keunsu Lyu, Hanbaek Kim, Jinsu Jung, Jae-Hun |
| contents | We propose a novel methodology for forecasting spatio-temporal data using supervised semi-nonnegative matrix factorization (SSNMF) with frequency regularization. Matrix factorization is employed to decompose spatio-temporal data into spatial and temporal components. To improve clarity in the temporal patterns, we introduce a nonnegativity constraint on the time domain along with regularization in the frequency domain. Specifically, regularization in the frequency domain involves selecting features in the frequency space, making an interpretation in the frequency domain more convenient. We propose two methods in the frequency domain: soft and hard regularizations, and provide convergence guarantees to first-order stationary points of the corresponding constrained optimization problem. While our primary motivation stems from geophysical data analysis based on GRACE (Gravity Recovery and Climate Experiment) data, our methodology has the potential for wider application. Consequently, when applying our methodology to GRACE data, we find that the results with the proposed methodology are comparable to previous research in the field of geophysical sciences but offer clearer interpretability. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2311_08636 |
| institution | arXiv |
| publishDate | 2023 |
| record_format | arxiv |
| spellingShingle | Supervised low-rank semi-nonnegative matrix factorization with frequency regularization for forecasting spatio-temporal data Kim, Keunsu Lyu, Hanbaek Kim, Jinsu Jung, Jae-Hun Machine Learning 65F22, 65F55 and 86A04 We propose a novel methodology for forecasting spatio-temporal data using supervised semi-nonnegative matrix factorization (SSNMF) with frequency regularization. Matrix factorization is employed to decompose spatio-temporal data into spatial and temporal components. To improve clarity in the temporal patterns, we introduce a nonnegativity constraint on the time domain along with regularization in the frequency domain. Specifically, regularization in the frequency domain involves selecting features in the frequency space, making an interpretation in the frequency domain more convenient. We propose two methods in the frequency domain: soft and hard regularizations, and provide convergence guarantees to first-order stationary points of the corresponding constrained optimization problem. While our primary motivation stems from geophysical data analysis based on GRACE (Gravity Recovery and Climate Experiment) data, our methodology has the potential for wider application. Consequently, when applying our methodology to GRACE data, we find that the results with the proposed methodology are comparable to previous research in the field of geophysical sciences but offer clearer interpretability. |
| title | Supervised low-rank semi-nonnegative matrix factorization with frequency regularization for forecasting spatio-temporal data |
| topic | Machine Learning 65F22, 65F55 and 86A04 |
| url | https://arxiv.org/abs/2311.08636 |