Large scale tensor regression using kernels and variational inference
We outline an inherent flaw of tensor factorization models when latent factors are expressed as a function of side information and propose a novel method to mitigate this. We coin our methodology Kernel Fried Tensor (KFT) and present it as a large-scale prediction and forecasting tool for high dimen...
| Main Authors: | , , |
|---|---|
| 格式: | Journal article |
| 語言: | English |
| 出版: |
Springer Nature
2021
|
| _version_ | 1826308473443319808 |
|---|---|
| author | Hu, R Nicholls, GK Sejdinovic, D |
| author_facet | Hu, R Nicholls, GK Sejdinovic, D |
| author_sort | Hu, R |
| collection | OXFORD |
| description | We outline an inherent flaw of tensor factorization models when latent
factors are expressed as a function of side information and propose a novel method
to mitigate this. We coin our methodology Kernel Fried Tensor (KFT) and present
it as a large-scale prediction and forecasting tool for high dimensional data. Our
results show superior performance against LightGBM and Field Aware Factorization
Machines (FFM), two algorithms with proven track records, widely used in largescale prediction. We also develop a variational inference framework for KFT
which enables associating the predictions and forecasts with calibrated uncertainty
estimates on several datasets. |
| first_indexed | 2024-03-07T07:20:03Z |
| format | Journal article |
| id | oxford-uuid:9743a5fd-3d87-4a03-87dc-a1e38c9f68a3 |
| institution | University of Oxford |
| language | English |
| last_indexed | 2024-03-07T07:20:03Z |
| publishDate | 2021 |
| publisher | Springer Nature |
| record_format | dspace |
| spelling | oxford-uuid:9743a5fd-3d87-4a03-87dc-a1e38c9f68a32022-09-27T14:50:00ZLarge scale tensor regression using kernels and variational inferenceJournal articlehttp://purl.org/coar/resource_type/c_dcae04bcuuid:9743a5fd-3d87-4a03-87dc-a1e38c9f68a3EnglishSymplectic ElementsSpringer Nature2021Hu, RNicholls, GKSejdinovic, DWe outline an inherent flaw of tensor factorization models when latent factors are expressed as a function of side information and propose a novel method to mitigate this. We coin our methodology Kernel Fried Tensor (KFT) and present it as a large-scale prediction and forecasting tool for high dimensional data. Our results show superior performance against LightGBM and Field Aware Factorization Machines (FFM), two algorithms with proven track records, widely used in largescale prediction. We also develop a variational inference framework for KFT which enables associating the predictions and forecasts with calibrated uncertainty estimates on several datasets. |
| spellingShingle | Hu, R Nicholls, GK Sejdinovic, D Large scale tensor regression using kernels and variational inference |
| title | Large scale tensor regression using kernels and variational inference |
| title_full | Large scale tensor regression using kernels and variational inference |
| title_fullStr | Large scale tensor regression using kernels and variational inference |
| title_full_unstemmed | Large scale tensor regression using kernels and variational inference |
| title_short | Large scale tensor regression using kernels and variational inference |
| title_sort | large scale tensor regression using kernels and variational inference |
| work_keys_str_mv | AT hur largescaletensorregressionusingkernelsandvariationalinference AT nichollsgk largescaletensorregressionusingkernelsandvariationalinference AT sejdinovicd largescaletensorregressionusingkernelsandvariationalinference |