Random tessellation forests
Space partitioning methods such as random forests and the Mondrian process are powerful machine learning methods for multi-dimensional and relational data, and are based on recursively cutting a domain. The flexibility of these methods is often limited by the requirement that the cuts be axis aligne...
| Auteurs principaux: | , , , , |
|---|---|
| Format: | Conference item |
| Langue: | English |
| Publié: |
Curran Associates
2019
|
| _version_ | 1826317654254682112 |
|---|---|
| author | Ge, S Wang, S Teh, YW Wang, L Elliott, LT |
| author_facet | Ge, S Wang, S Teh, YW Wang, L Elliott, LT |
| author_sort | Ge, S |
| collection | OXFORD |
| description | Space partitioning methods such as random forests and the Mondrian process are powerful machine learning methods for multi-dimensional and relational data, and are based on recursively cutting a domain. The flexibility of these methods is often limited by the requirement that the cuts be axis aligned. The Ostomachion process and the self-consistent binary space partitioning-tree process were recently introduced as generalizations of the Mondrian process for space partitioning with non-axis aligned cuts in the plane. Motivated by the need for a multi-dimensional partitioning tree with non-axis aligned cuts, we propose the Random Tessellation Process, a framework that includes the Mondrian process as a special case. We derive a sequential Monte Carlo algorithm for inference, and provide random forest methods. Our methods are self-consistent and can relax axis-aligned constraints, allowing complex inter-dimensional dependence to be captured. We present a simulation study and analyze gene expression data of brain tissue, showing improved accuracies over other methods. |
| first_indexed | 2024-03-07T04:08:14Z |
| format | Conference item |
| id | oxford-uuid:c6f16fca-986c-4be7-b1a6-d3b1beb0dd9d |
| institution | University of Oxford |
| language | English |
| last_indexed | 2025-03-11T16:57:20Z |
| publishDate | 2019 |
| publisher | Curran Associates |
| record_format | dspace |
| spelling | oxford-uuid:c6f16fca-986c-4be7-b1a6-d3b1beb0dd9d2025-02-21T12:53:10ZRandom tessellation forestsConference itemhttp://purl.org/coar/resource_type/c_5794uuid:c6f16fca-986c-4be7-b1a6-d3b1beb0dd9dEnglishSymplectic Elements Curran Associates2019Ge, SWang, STeh, YWWang, LElliott, LTSpace partitioning methods such as random forests and the Mondrian process are powerful machine learning methods for multi-dimensional and relational data, and are based on recursively cutting a domain. The flexibility of these methods is often limited by the requirement that the cuts be axis aligned. The Ostomachion process and the self-consistent binary space partitioning-tree process were recently introduced as generalizations of the Mondrian process for space partitioning with non-axis aligned cuts in the plane. Motivated by the need for a multi-dimensional partitioning tree with non-axis aligned cuts, we propose the Random Tessellation Process, a framework that includes the Mondrian process as a special case. We derive a sequential Monte Carlo algorithm for inference, and provide random forest methods. Our methods are self-consistent and can relax axis-aligned constraints, allowing complex inter-dimensional dependence to be captured. We present a simulation study and analyze gene expression data of brain tissue, showing improved accuracies over other methods. |
| spellingShingle | Ge, S Wang, S Teh, YW Wang, L Elliott, LT Random tessellation forests |
| title | Random tessellation forests |
| title_full | Random tessellation forests |
| title_fullStr | Random tessellation forests |
| title_full_unstemmed | Random tessellation forests |
| title_short | Random tessellation forests |
| title_sort | random tessellation forests |
| work_keys_str_mv | AT ges randomtessellationforests AT wangs randomtessellationforests AT tehyw randomtessellationforests AT wangl randomtessellationforests AT elliottlt randomtessellationforests |