Recommendations for Baselines and Benchmarking Approximate Gaussian Processes
Fuente:
arXiv
Saved in:
| Main Authors: | , , , , |
|---|---|
| Format: | Preprint |
| Published: |
2024
|
| Subjects: | |
| Online Access: | |
| Tags: |
Add Tag
No Tags, Be the first to tag this record!
|
| _version_ | 1866914679867047936 |
|---|---|
| author | Ober, Sebastian W. Artemev, Artem Wagenländer, Marcel Grobins, Rudolfs van der Wilk, Mark |
| author_facet | Ober, Sebastian W. Artemev, Artem Wagenländer, Marcel Grobins, Rudolfs van der Wilk, Mark |
| contents | Gaussian processes (GPs) are a mature and widely-used component of the ML toolbox. One of their desirable qualities is automatic hyperparameter selection, which allows for training without user intervention. However, in many realistic settings, approximations are typically needed, which typically do require tuning. We argue that this requirement for tuning complicates evaluation, which has led to a lack of a clear recommendations on which method should be used in which situation. To address this, we make recommendations for comparing GP approximations based on a specification of what a user should expect from a method. In addition, we develop a training procedure for the variational method of Titsias [2009] that leaves no choices to the user, and show that this is a strong baseline that meets our specification. We conclude that benchmarking according to our suggestions gives a clearer view of the current state of the field, and uncovers problems that are still open that future papers should address. |
| format | Preprint |
| id |
arxiv_https___arxiv_org_abs_2402_09849 |
| institution | arXiv |
| publishDate | 2024 |
| record_format | arxiv |
| spellingShingle | Recommendations for Baselines and Benchmarking Approximate Gaussian Processes Ober, Sebastian W. Artemev, Artem Wagenländer, Marcel Grobins, Rudolfs van der Wilk, Mark Machine Learning Gaussian processes (GPs) are a mature and widely-used component of the ML toolbox. One of their desirable qualities is automatic hyperparameter selection, which allows for training without user intervention. However, in many realistic settings, approximations are typically needed, which typically do require tuning. We argue that this requirement for tuning complicates evaluation, which has led to a lack of a clear recommendations on which method should be used in which situation. To address this, we make recommendations for comparing GP approximations based on a specification of what a user should expect from a method. In addition, we develop a training procedure for the variational method of Titsias [2009] that leaves no choices to the user, and show that this is a strong baseline that meets our specification. We conclude that benchmarking according to our suggestions gives a clearer view of the current state of the field, and uncovers problems that are still open that future papers should address. |
| title | Recommendations for Baselines and Benchmarking Approximate Gaussian Processes |
| topic | Machine Learning |
| url | https://arxiv.org/abs/2402.09849 |