Computer Science > Machine Learning
[Submitted on 1 Aug 2018 (v1), last revised 26 Jul 2023 (this version, v9)]
Title:Compressible Spectral Mixture Kernels with Sparse Dependency Structures for Gaussian Processes
View PDFAbstract:Spectral mixture (SM) kernels comprise a powerful class of generalized kernels for Gaussian processes (GPs) to describe complex patterns. This paper introduces model compression and time- and phase (TP) modulated dependency structures to the original (SM) kernel for improved generalization of GPs. Specifically, by adopting Bienaymés identity, we generalize the dependency structure through cross-covariance between the SM components. Then, we propose a novel SM kernel with a dependency structure (SMD) by using cross-convolution between the SM components. Furthermore, we ameliorate the expressiveness of the dependency structure by parameterizing it with time and phase delays. The dependency structure has clear interpretations in terms of spectral density, covariance behavior, and sampling path. To enrich the SMD with effective hyperparameter initialization, compressible SM kernel components, and sparse dependency structures, we introduce a novel structure adaptation (SA) algorithm in the end. A thorough comparative analysis of the SMD on both synthetic and real-life applications corroborates its efficacy.
Submission history
From: Kai Chen [view email][v1] Wed, 1 Aug 2018 20:55:54 UTC (1,056 KB)
[v2] Sun, 9 Sep 2018 11:50:23 UTC (2,524 KB)
[v3] Thu, 13 Sep 2018 21:37:31 UTC (2,524 KB)
[v4] Tue, 18 Sep 2018 09:05:19 UTC (2,676 KB)
[v5] Sun, 14 Oct 2018 20:26:09 UTC (2,606 KB)
[v6] Fri, 16 Aug 2019 19:18:41 UTC (5,880 KB)
[v7] Tue, 10 Aug 2021 02:09:14 UTC (1,366 KB)
[v8] Tue, 31 Aug 2021 12:23:05 UTC (1,366 KB)
[v9] Wed, 26 Jul 2023 04:30:49 UTC (1,216 KB)
Current browse context:
cs.LG
References & Citations
Bibliographic and Citation Tools
Bibliographic Explorer (What is the Explorer?)
Connected Papers (What is Connected Papers?)
Litmaps (What is Litmaps?)
scite Smart Citations (What are Smart Citations?)
Code, Data and Media Associated with this Article
alphaXiv (What is alphaXiv?)
CatalyzeX Code Finder for Papers (What is CatalyzeX?)
DagsHub (What is DagsHub?)
Gotit.pub (What is GotitPub?)
Hugging Face (What is Huggingface?)
Papers with Code (What is Papers with Code?)
ScienceCast (What is ScienceCast?)
Demos
Recommenders and Search Tools
Influence Flower (What are Influence Flowers?)
CORE Recommender (What is CORE?)
IArxiv Recommender
(What is IArxiv?)
arXivLabs: experimental projects with community collaborators
arXivLabs is a framework that allows collaborators to develop and share new arXiv features directly on our website.
Both individuals and organizations that work with arXivLabs have embraced and accepted our values of openness, community, excellence, and user data privacy. arXiv is committed to these values and only works with partners that adhere to them.
Have an idea for a project that will add value for arXiv's community? Learn more about arXivLabs.