#2541 closed feature request (done)
Student-t Process
| Reported by: | gkronber | Owned by: | gkronber |
|---|---|---|---|
| Priority: | medium | Milestone: | |
| Component: | Algorithms.DataAnalysis | Version: | 3.3.13 |
| Keywords: | Cc: |
Description
See "Student-t Processes as Alternatives to Gaussian Processes", Amar Shah, Andrew Gordon Wilson, Zoubin Ghahramani, 2014
http://arxiv.org/abs/1402.4306
Could be implemented rather easily using the existing framework for Gaussian Processes
Change History (11)
comment:1 Changed 8 years ago by gkronber
- Milestone changed from HeuristicLab 4.x Backlog to HeuristicLab 3.3.14
- Owner set to gkronber
- Status changed from new to accepted
comment:2 Changed 8 years ago by gkronber
comment:3 Changed 8 years ago by gkronber
TODO:
- improve stability for covariance matrix decomposition (special handling of noise covariance parameter)
- fix gradient calculation for mean function
comment:4 Changed 8 years ago by gkronber
comment:5 Changed 8 years ago by gkronber
Changes in this ticket should be undone before the release. Reasoning:
- Code-duplication
- Implementation is not stable enough yet
- Functionality has not been very useful so far.
comment:6 Changed 8 years ago by gkronber
r13975: deleted implementation of StudentT process
comment:7 Changed 8 years ago by gkronber
comment:8 Changed 8 years ago by gkronber
comment:9 Changed 8 years ago by gkronber
- Status changed from accepted to readytorelease
comment:10 Changed 8 years ago by mkommend
- Resolution set to done
- Status changed from readytorelease to closed
comment:11 Changed 8 years ago by gkronber
- Milestone HeuristicLab 3.3.14 deleted
Note: See
TracTickets for help on using
tickets.
r13438: added crude implementation of Student-t process (using almost the same source code as GP model)