Scalable Inference for Gaussian Process Models with Black-Box Likelihoods

By Amir Dezfouli, Edwin V. Bonilla. 2015

68 citations recorded by the source corpus.

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

Citations in the corpus

Functional Regularisation for Continual Learning using Gaussian Processes (2019, methodology): “In the non-conjugate generalisations of this method (Hensman et al., 2013; Lloyd et al., 2015; Dezfouli & Bonilla, 2015; Hensman et al., 2015; Sheth et al., 2015), the lower bound is written as N1∑ j=1 Eq(f1,j)[log p(y1,j |f1,j)]− KL(q(u1)||pθ(u1)), (5) where q(f1,j) = ∫ p(f1,j |u1)q(u1)du1 is an…”
Stochastic Variational Deep Kernel Learning (2016, methodology): “In particular, a similar variational lower bound as in Eq (7) was proposed in [25, 11] for a sparse GP, which were extended to non-conjugate likelihoods, with the intractable integrals estimated using Gaussian quadrature as in the KLSP-GP [12] or univariate Gaussian samples as in the SAVI-GP [7].”
Streaming Sparse Gaussian Process Approximations (2017, background): “Stochastic optimisation methods that employ mini-batches of training data can be used to further reduce computational costs [4, 5, 6, 7], allowing GPs to be scaled to datasets comprising millions of data points.”
Continual Multi-task Gaussian Processes (2019, methodology): “Besides popular scalable GP methods for dealing with non-Gaussian likelihoods (Dezfouli and Bonilla, 2015; Hensman et al., 2015) in both single and multi-output scenarios, we focus in the open problem of heterogeneous likelihood models.”
A Unifying Framework for Gaussian Process Pseudo-Point Approximations using Power Expectation Propagation (2016, methodology): “Since the proposed Power EP approach is general, an extension to other likelihood functions is as simple as for VFE methods (Dezfouli and Bonilla, 2015).”
A Unifying Framework for Sparse Gaussian Process Approximation using Power Expectation Propagation (2016, methodology): “or GP classication is recovered (Rasmussen and Williams, 2005, sec. 3.6). Since the proposed Power EP approach is general, an extension to other likelihood functions is as simple as for VFE methods (Dezfouli and Bonilla, 2015). For example, the multinomial probit likelihood can be handled in the sam”