It is all in the noise: Efficient multi-task Gaussian process inference with structured residuals
Multi-task prediction models are widely being used to couple regressors or classification models by sharing information across related tasks. A common pitfall of these models is that they assume that the output tasks are independent conditioned on the inputs. Here, we propose a multi-task Gaussian process approach to model both the relatedness between regressors as well as the task correlations in the residuals, in order to more accurately identify true sharing between regressors. The resulting Gaussian model has a covariance term that is the sum of Kronecker products, for which efficient parameter inference and out of sample prediction are feasible. On both synthetic examples and applications to phenotype prediction in genetics, we find substantial benefits of modeling structured noise compared to established alternatives.
Code (0)
등록된 구현이 없습니다.
Tasks
AllGeneral ClassificationPredictionMethods 이 논문이 사용한 방법론
Similar Papers 제목 키워드 기반
Discrete-Time Diffusion-Like Models for Speech Synthesis
Diffusion models have attracted a lot of attention in recent years. These models view speech generation as a continuous-time process. For efficient training, this process is typically restricted to additive Gaussian nois…
Speech SynthesisRobust Multi-Class Gaussian Process Classification
Multi-class Gaussian Process Classifiers (MGPCs) are often affected by over-fitting problems when labeling errors occur far from the decision boundaries. To prevent this, we investigate a robust MGPC (RMGPC) which consid…
ClassificationGeneral ClassificationGaussian Process Regression Networks
We introduce a new regression framework, Gaussian process regression networks (GPRN), which combines the structural properties of Bayesian neural networks with the non-parametric flexibility of Gaussian processes. This m…
Gaussian ProcessesregressionScalable Gaussian Process Classification with Additive Noise for Various Likelihoods
Gaussian process classification (GPC) provides a flexible and powerful statistical framework describing joint distributions over function space. Conventional GPCs however suffer from (i) poor scalability for big data due…
ClassificationGeneral ClassificationGPRMulti-class Classification+1Deep Gaussian Processes with Importance-Weighted Variational Inference
Deep Gaussian processes (DGPs) can model complex marginal densities as well as complex mappings. Non-Gaussian marginals are essential for modelling real-world data, and can be generated from the DGP by incorporating unco…
Gaussian ProcessesVariational Inference