arXiv subjects

James Hensman

Publications and source records attributed to James Hensman.

52 records · Page 3Linked to original sources

arXiv2017

Scalable Joint Models for Reliable Uncertainty-Aware Event Prediction

Missing data and noisy observations pose significant challenges for reliably predicting events from irregularly sampled multivariate time series (longitudinal) data. Imputation methods, which are typically used for completing the data prior to event prediction, lack a principled mechanism to account for the uncertainty due to missingness. Alternatively, state-of-the-art joint modeling techniques can be used for jointly modeling the longitudinal and event data and compute event probabilities conditioned on the longitudinal observations. These approaches, however, make strong parametric assumptions and do not easily scale to multivariate signals with many observations. Our proposed approach consists of several key innovations. First, we develop a flexible and scalable joint model based upon sparse multiple-output Gaussian processes. Unlike state-of-the-art joint models, the proposed model can explain highly challenging structure including non-Gaussian noise while scaling to large data. Second, we derive an optimal policy for predicting events using the distribution of the event occurrence estimated by the joint model. The derived policy trades-off the cost of a delayed detection versus incorrect assessments and abstains from making decisions when the estimated event probability does not satisfy the derived confidence criteria. Experiments on a large dataset show that the proposed framework significantly outperforms state-of-the-art techniques in event prediction.

stat.ML↗

arXiv2016

GPflow: A Gaussian process library using TensorFlow

GPflow is a Gaussian process library that uses TensorFlow for its core computations and Python for its front end. The distinguishing features of GPflow are that it uses variational inference as the primary approximation method, provides concise code through the use of automatic differentiation, has been engineered with a particular emphasis on software testing and is able to exploit GPU hardware.

stat.ML↗

arXiv2016

Gaussian process models for periodicity detection

We consider the problem of detecting and quantifying the periodic component of a function given noise-corrupted observations of a limited number of input/output tuples. Our approach is based on Gaussian process regression which provides a flexible non-parametric framework for modelling periodic data. We introduce a novel decomposition of the covariance function as the sum of periodic and aperiodic kernels. This decomposition allows for the creation of sub-models which capture the periodic nature of the signal and its complement. To quantify the periodicity of the signal, we derive a periodicity ratio which reflects the uncertainty in the fitted sub-models. Although the method can be applied to many kernels, we give a special emphasis to the Matérn family, from the expression of the reproducing kernel Hilbert space inner product to the implementation of the associated periodic kernels in a Gaussian process toolkit. The proposed method is illustrated by considering the detection of periodically expressed genes in the arabidopsis genome.

math.ST↗

arXiv2016

Chained Gaussian Processes

Gaussian process models are flexible, Bayesian non-parametric approaches to regression. Properties of multivariate Gaussians mean that they can be combined linearly in the manner of additive models and via a link function (like in generalized linear models) to handle non-Gaussian data. However, the link function formalism is restrictive, link functions are always invertible and must convert a parameter of interest to a linear combination of the underlying processes. There are many likelihoods and models where a non-linear combination is more appropriate. We term these more general models Chained Gaussian Processes: the transformation of the GPs to the likelihood parameters will not generally be invertible, and that implies that linearisation would only be possible with multiple (localized) links, i.e. a chain. We develop an approximate inference procedure for Chained GPs that is scalable and applicable to any factorized likelihood. We demonstrate the approximation on a range of likelihood functions.

stat.ML↗

arXiv2015

On Sparse variational methods and the Kullback-Leibler divergence between stochastic processes

The variational framework for learning inducing variables (Titsias, 2009a) has had a large impact on the Gaussian process literature. The framework may be interpreted as minimizing a rigorously defined Kullback-Leibler divergence between the approximating and posterior processes. To our knowledge this connection has thus far gone unremarked in the literature. In this paper we give a substantial generalization of the literature on this topic. We give a new proof of the result for infinite index sets which allows inducing points that are not data points and likelihoods that depend on all function values. We then discuss augmented index sets and show that, contrary to previous works, marginal consistency of augmentation is not enough to guarantee consistency of variational inference with the original model. We then characterize an extra condition where such a guarantee is obtainable. Finally we show how our framework sheds light on interdomain sparse approximations and sparse approximations for Cox processes.

stat.ML↗

arXiv2015

ABC for climate: dealing with expensive simulators

This paper is due to appear as a chapter of the forthcoming Handbook of Approximate Bayesian Computation (ABC) by S. Sisson, L. Fan, and M. Beaumont. We describe the challenge of calibrating climate simulators, and discuss the differences in emphasis in climate science compared to many of the more traditional ABC application areas. The primary difficulty is how to do inference with a computationally expensive simulator which we can only afford to run a small number of times, and we describe how Gaussian process emulators are used as surrogate models in this case. We introduce the idea of history matching, which is a non-probabilistic calibration method, which divides the parameter space into (not im)plausible and implausible regions. History matching can be shown to be a special case of ABC, but with a greater emphasis on defining realistic simulator discrepancy bounds, and using these to define tolerances and metrics. We describe a design approach for choosing parameter values at which to run the simulator, and illustrate the approach on a toy climate model, showing that with careful design we can find the plausible region with a very small number of model evaluations. Finally, we describe how calibrated GENIE-1 (an earth system model of intermediate complexity) predictions have been used, and why it is important to accurately characterise parametric uncertainty.

stat.CO↗

arXiv2015

Fast and accurate approximate inference of transcript expression from RNA-seq data

Motivation: Assigning RNA-seq reads to their transcript of origin is a fundamental task in transcript expression estimation. Where ambiguities in assignments exist due to transcripts sharing sequence, e.g. alternative isoforms or alleles, the problem can be solved through probabilistic inference. Bayesian methods have been shown to provide accurate transcript abundance estimates compared to competing methods. However, exact Bayesian inference is intractable and approximate methods such as Markov chain Monte Carlo (MCMC) and Variational Bayes (VB) are typically used. While providing a high degree of accuracy and modelling flexibility, standard implementations can be prohibitively slow for large datasets and complex transcriptome annotations. Results: We propose a novel approximate inference scheme based on VB and apply it to an existing model of transcript expression inference from RNA-seq data. Recent advances in VB algorithmics are used to improve the convergence of the algorithm beyond the standard Variational Bayes Expectation Maximisation (VBEM) algorithm. We apply our algorithm to simulated and biological datasets, demonstrating a significant increase in speed with only very small loss in accuracy of expression level estimation. We carry out a comparative study against seven popular alternative methods and demonstrate that our new algorithm provides excellent accuracy and inter-replicate consistency while remaining competitive in computation time. Availability: The methods were implemented in R and C++, and are available as part of the BitSeq project at \url{https://github.com/BitSeq}. The method is also available through the BitSeq Bioconductor package. The source code to reproduce all simulation results can be accessed via \url{https://github.com/BitSeq/BitSeqVB_benchmarking}.

q-bio.QM↗

arXiv2015

MCMC for Variationally Sparse Gaussian Processes

Gaussian process (GP) models form a core part of probabilistic machine learning. Considerable research effort has been made into attacking three issues with GP models: how to compute efficiently when the number of data is large; how to approximate the posterior when the likelihood is not Gaussian and how to estimate covariance function parameter posteriors. This paper simultaneously addresses these, using a variational approximation to the posterior which is sparse in support of the function but otherwise free-form. The result is a Hybrid Monte-Carlo sampling scheme which allows for a non-Gaussian approximation over the function values and covariance parameters simultaneously, with efficient computations based on inducing-point sparse GPs. Code to replicate each experiment in this paper will be available shortly.

stat.ML↗

arXiv2015

Spike and Slab Gaussian Process Latent Variable Models

The Gaussian process latent variable model (GP-LVM) is a popular approach to non-linear probabilistic dimensionality reduction. One design choice for the model is the number of latent variables. We present a spike and slab prior for the GP-LVM and propose an efficient variational inference procedure that gives a lower bound of the log marginal likelihood. The new model provides a more principled approach for selecting latent dimensions than the standard way of thresholding the length-scale parameters. The effectiveness of our approach is demonstrated through experiments on real and simulated data. Further, we extend multi-view Gaussian processes that rely on sharing latent dimensions (known as manifold relevance determination) with spike and slab priors. This allows a more principled approach for selecting a subset of the latent space for each view of data. The extended model outperforms the previous state-of-the-art when applied to a cross-modal multimedia retrieval task.

stat.ML↗

arXiv2015

Fast Approximate Inference of Transcript Expression Levels from RNA-seq Data

Motivation: The mapping of RNA-seq reads to their transcripts of origin is a fundamental task in transcript expression estimation and differential expression scoring. Where ambiguities in mapping exist due to transcripts sharing sequence, e.g. alternative isoforms or alleles, the problem becomes an instance of non-trivial probabilistic inference. Bayesian inference in such a problem is intractable and approximate methods must be used such as Markov chain Monte Carlo (MCMC) and Variational Bayes. Standard implementations of these methods can be prohibitively slow for large datasets and complex gene models. Results: We propose an approximate inference scheme based on Variational Bayes applied to an existing model of transcript expression inference from RNA-seq data. We apply recent advances in Variational Bayes algorithmics to improve the convergence of the algorithm beyond the standard variational expectation-maximisation approach. We apply our algorithm to simulated and biological datasets, demonstrating that the increase in speed requires only a small trade-off in accuracy of expression level estimation. Availability: The methods were implemented in R and C++, and are available as part of the BitSeq project at https://code.google.com/p/bitseq/. The methods will be made available through the BitSeq Bioconductor package at the next stable release.

q-bio.GN↗

arXiv2014

Nested Variational Compression in Deep Gaussian Processes

Deep Gaussian processes provide a flexible approach to probabilistic modelling of data using either supervised or unsupervised learning. For tractable inference approximations to the marginal likelihood of the model must be made. The original approach to approximate inference in these models used variational compression to allow for approximate variational marginalization of the hidden variables leading to a lower bound on the marginal likelihood of the model [Damianou and Lawrence, 2013]. In this paper we extend this idea with a nested variational compression. The resulting lower bound on the likelihood can be easily parallelized or adapted for stochastic variational inference.

stat.ML↗

arXiv2014

Scalable Variational Gaussian Process Classification

Gaussian process classification is a popular method with a number of appealing properties. We show how to scale the model within a variational inducing point framework, outperforming the state of the art on benchmark datasets. Importantly, the variational formulation can be exploited to allow classification in problems with millions of data points, as we demonstrate in experiments.

stat.ML↗

arXiv2014

Gaussian Process Models with Parallelization and GPU acceleration

In this work, we present an extension of Gaussian process (GP) models with sophisticated parallelization and GPU acceleration. The parallelization scheme arises naturally from the modular computational structure w.r.t. datapoints in the sparse Gaussian process formulation. Additionally, the computational bottleneck is implemented with GPU acceleration for further speed up. Combining both techniques allows applying Gaussian process models to millions of datapoints. The efficiency of our algorithm is demonstrated with a synthetic dataset. Its source code has been integrated into our popular software library GPy.

cs.DC↗

arXiv2014

Fast nonparametric clustering of structured time-series

In this publication, we combine two Bayesian non-parametric models: the Gaussian Process (GP) and the Dirichlet Process (DP). Our innovation in the GP model is to introduce a variation on the GP prior which enables us to model structured time-series data, i.e. data containing groups where we wish to model inter- and intra-group variability. Our innovation in the DP model is an implementation of a new fast collapsed variational inference procedure which enables us to optimize our variationala pproximation significantly faster than standard VB approaches. In a biological time series application we show how our model better captures salient features of the data, leading to better consistency with existing biological classifications, while the associated inference algorithm provides a twofold speed-up over EM-based variational inference.

cs.LG↗

arXiv2013

Gaussian Processes for Big Data

We introduce stochastic variational inference for Gaussian process models. This enables the application of Gaussian process (GP) models to data sets containing millions of data points. We show how GPs can be vari- ationally decomposed to depend on a set of globally relevant inducing variables which factorize the model in the necessary manner to perform variational inference. Our ap- proach is readily extended to models with non-Gaussian likelihoods and latent variable models based around Gaussian processes. We demonstrate the approach on a simple toy problem and two real world data sets.

cs.LG↗

arXiv2012

Fast Variational Inference in the Conjugate Exponential Family

We present a general method for deriving collapsed variational inference algo- rithms for probabilistic models in the conjugate exponential family. Our method unifies many existing approaches to collapsed variational inference. Our collapsed variational inference leads to a new lower bound on the marginal likelihood. We exploit the information geometry of the bound to derive much faster optimization methods based on conjugate gradients for these models. Our approach is very general and is easily applied to any model where the mean field update equations have been derived. Empirically we show significant speed-ups for probabilistic models optimized using our bound.

cs.LG↗