Tree-structured Gaussian process approximations Supplementary material
نویسندگان
چکیده
subject to q(f |u) = ∏ i q(fi|u) and ∫ dfiq(fi|u) = 1. It is noted that KL(a||b) is the measurement of information “lost” when using b to approximate a. It was argued in [1] that it is appropriate to use this KL divergence as an approximation measure since we are trying to find a sparse representation u and its relationship with f to approximate p by q. The KL divergence above can be expanded as follows, L1 = KL(p(f ,u)||q(f ,u)) (2)
منابع مشابه
Supplementary material:Bayesian Optimization with Tree-structured Dependencies
In this supplementary material, we provide additional experimental results along with some details about the inference in our model and its nonlinear extension. 1. Experiments In this section, we provide complementary experimental results. 1.1. Optimization of synthetic tree-structured functions In addition to the results presented in the core of the paper, we include simulations with the non-l...
متن کاملGray-box Inference for Structured Gaussian Process Models: Supplementary Material
1 Proof of Theorem 2 Here we proof the result that we can estimate the expected log likelihood and its gradients using expectations over low-dimensional Gaussians, that is that Theorem. For the structured gp model defined in our paper the expected log likelihood over the given variational distribution and its gradients can be estimated using expectations over Tn-dimensional Gaussians and |V|-di...
متن کاملPerturbative corrections for approximate inference in Gaussian latent variable models
Expectation Propagation (EP) provides a framework for approximate inference. When the model under consideration is over a latent Gaussian field, with the approximation being Gaussian, we show how these approximations can systematically be corrected. A perturbative expansion is made of the exact but intractable correction, and can be applied to the model’s partition function and other moments of...
متن کاملTree-structured Gaussian Process Approximations
Gaussian process regression can be accelerated by constructing a small pseudodataset to summarize the observed data. This idea sits at the heart of many approximation schemes, but such an approach requires the number of pseudo-datapoints to be scaled with the range of the input space if the accuracy of the approximation is to be maintained. This presents problems in time-series settings or in s...
متن کاملGaussian Process Regression Networks Supplementary Material
In this supplementary material, we discuss some further details of our ESS and VB inference (Sections 1 and 2), the computational complexity of our inference procedures (Section 3), and the correlation structure induced by the GPRN model (Section 4). We also discuss multimodality in the GPRN posterior (Section 5), SVLMC, and some background information and notation for Gaussian process regressi...
متن کامل