Scientific experiments are usually expensive due to complex experimental preparation and processing. Experimental design is therefore involved with the task of finding the optimal experimental input that results in the desirable output by using as few experiments as possible. Experimenters can often acquire the knowledge about the location of the global optimum. However, they do not know how to exploit this knowledge to accelerate experimental design. In this paper, we adopt the technique of Bayesian optimization for experimental design since Bayesian optimization has established itself as an efficient tool for optimizing expensive black-box functions. Again, it is unknown how to incorporate the expert prior knowledge about the global optimum into Bayesian optimization process. To address it, we represent the expert knowledge about the global optimum via placing a prior distribution on it and we then derive its posterior distribution. An efficient Bayesian optimization approach has been proposed via posterior sampling on the posterior distribution of the global optimum. We theoretically analyze the convergence of the proposed algorithm and discuss the robustness of incorporating expert prior. We evaluate the efficiency of our algorithm by optimizing synthetic functions and tuning hyperparameters of classifiers along with a real-world experiment on the synthesis of short polymer fiber. The results clearly demonstrate the advantages of our proposed method.
Imperial College London（帝国理工学院）；University of Bristol（布里斯托大学）；University of Cambridge（剑桥大学）；ETH Zürich（苏黎世联邦理工学院）
Bayesian neural networks that incorporate data augmentation implicitly use a ``randomly perturbed log-likelihood [which] does not have a clean interpretation as a valid likelihood function'' (Izmailov et al. 2021). Here, we provide several approaches to developing principled Bayesian neural networks incorporating data augmentation. We introduce a ``finite orbit'' setting which allows likelihoods to be computed exactly, and give tight multi-sample bounds in the more usual ``full orbit'' setting. These models cast light on the origin of the cold posterior effect. In particular, we find that the …
TU Graz（格拉茨工业大学）；OFAI（奥地利人工智能研究所）；TU Eindhoven（埃因霍温理工大学）；University of Cambridge（剑桥大学）
Gaussian Processes (GPs) are powerful non-parametric Bayesian regression models that allow exact posterior inference, but exhibit high computational and memory costs. In order to improve scalability of GPs, approximate posterior inference is frequently employed, where a prominent class of approximation techniques is based on local GP experts. However, local-expert techniques proposed so far are either not well-principled, come with limited approximation guarantees, or lead to intractable models. In this paper, we introduce deep structured mixtures of GP experts, a stochastic process model whic…
In order to estimate model parameters and circumvent possible difficulties encountered with the likelihood function, we propose to replace the likelihood in the formula of the posterior distribution by a function depending on a contrast. The properties of the contrast-based (CB) posterior distribution and MAP estimator are studied to understand what the consequences of incorporating a contrast in the Bayesian formula are. We show that the proposed method can be used to make frequentist inference and allows the reduction of analytical calculations to get the limit variance matrix of the estimat…
A well-designed framework for risk classification and ratemaking in automobile insurance is key to insurers' profitability and risk management, while also ensuring that policyholders are charged a fair premium according to their risk profile. In this paper, we propose to adapt a flexible regression model, called the Mixed LRMoE, to the problem of a posteriori risk classification and ratemaking, where policyholder-level random effects are incorporated to better infer their risk profile reflected by the claim history. We also develop a stochastic variational Expectation-Conditional-Maximization …
Bayesian optimisation is a well-known sample-efficient method for the optimisation of expensive black-box functions. However when dealing with big search spaces the algorithm goes through several low function value regions before reaching the optimum of the function. Since the function evaluations are expensive in terms of both money and time, it may be desirable to alleviate this problem. One approach to subside this cold start phase is to use prior knowledge that can accelerate the optimisation. In its standard form, Bayesian optimisation assumes the likelihood of any point in the search spa…
Harvard University（哈佛大学）；University of Washington（华盛顿大学）；Google Brain（谷歌大脑）
We propose a black-box variational inference method to approximate intractable distributions with an increasingly rich approximating class. Our method, termed variational boosting, iteratively refines an existing variational approximation by solving a sequence of optimization problems, allowing the practitioner to trade computation time for accuracy. We show how to expand the variational approximating class by incorporating additional covariance structure and by introducing new components to form a mixture. We apply variational boosting to synthetic and real statistical models, and show that r…
Nanjing University of Posts and Telecommunications（南京邮电大学）
Bayesian optimization (BO) is a powerful paradigm for derivative-free global optimization of a black-box objective function (BOF) that is expensive to evaluate. However, the overhead of BO can still be prohibitive for problems with highly expensive function evaluations. In this paper, we investigate how to reduce the required number of function evaluations for BO without compromise in solution quality. We explore the idea of posterior regularization to harness low fidelity (LF) data within the Gaussian process upper confidence bound (GP-UCB) framework. The LF data can arise from previous evalu…
Austrian Research Institute for Artificial Intelligence（奥地利人工智能研究所）；Graz University of Technology（格拉茨工业大学）；University of Cambridge（剑桥大学）
While Gaussian processes (GPs) are the method of choice for regression tasks, they also come with practical difficulties, as inference cost scales cubic in time and quadratic in memory. In this paper, we introduce a natural and expressive way to tackle these problems, by incorporating GPs in sum-product networks (SPNs), a recently proposed tractable probabilistic model allowing exact and efficient inference. In particular, by using GPs as leaves of an SPN we obtain a novel flexible prior over functions, which implicitly represents an exponentially large mixture of local GPs. Exact and efficien…
In order to improve forecasts, a decisionmaker often combines probabilities given by various sources, such as human experts and machine learning classifiers. When few training data are available, aggregation can be improved by incorporating prior knowledge about the event being forecasted and about salient properties of the experts. To this end, we develop a generative Bayesian aggregation model for probabilistic classi cation. The model includes an event-specific prior, measures of individual experts' bias, calibration, accuracy, and a measure of dependence betweeen experts. Rather than requi…
Domains where supervised models are deployed often come with task-specific constraints, such as prior expert knowledge on the ground-truth function, or desiderata like safety and fairness. We introduce a novel probabilistic framework for reasoning with such constraints and formulate a prior that enables us to effectively incorporate them into Bayesian neural networks (BNNs), including a variant that can be amortized over tasks. The resulting Output-Constrained BNN (OC-BNN) is fully consistent with the Bayesian framework for uncertainty quantification and is amenable to black-box inference. Unl…
University of California, Los Angeles（加州大学洛杉矶分校）；David Geffen School of Medicine, University of California, Los Angeles（加州大学洛杉矶分校大卫·格芬医学院）
We develop a personalized real time risk scoring algorithm that provides timely and granular assessments for the clinical acuity of ward patients based on their (temporal) lab tests and vital signs. Heterogeneity of the patients population is captured via a hierarchical latent class model. The proposed algorithm aims to discover the number of latent classes in the patients population, and train a mixture of Gaussian Process (GP) experts, where each expert models the physiological data streams associated with a specific class. Self-taught transfer learning is used to transfer the knowledge of l…
KTH Royal Institute of Technology（瑞典皇家理工学院）；RaySearch Laboratories（瑞镭实验室）；Silo AI（Silo AI（西洛人工智能公司））
We present a new nonparametric mixture-of-experts model for multivariate regression problems, inspired by the probabilistic k-nearest neighbors algorithm. Using a conditionally specified model, predictions for out-of-sample inputs are based on similarities to each observed data point, yielding predictive distributions represented by Gaussian mixtures. Posterior inference is performed on the parameters of the mixture components as well as the distance metric using a mean-field variational Bayes algorithm accompanied with a stochastic gradient-based optimization procedure. The proposed method is…
Université du Québec à Montréal（蒙特利尔魁北克大学）；Université de Montréal（蒙特利尔大学）；Université catholique de Louvain（天主教鲁汶大学）
A recommender system based on ranks is proposed, where an expert's ranking of a set of objects and a user's ranking of a subset of those objects are combined to make a prediction of the user's ranking of all objects. The rankings are assumed to be induced by latent continuous variables corresponding to the grades assigned by the expert and the user to the objects. The dependence between the expert and user grades is modelled by a copula in some parametric family. Given a prior distribution on the copula parameter, the user's complete ranking is predicted by the mode of the posterior predictive…
TU Darmstadt（达姆施塔特工业大学）；Nanjing University of Aeronautics and Astronautics（南京航空航天大学）；Aalto University（阿尔托大学）；Centre for Cognitive Science TU Darmstadt（达姆施塔特工业大学认知科学中心）；Hessian Center for AI (hessian.AI)（黑森人工智能中心（hessian.AI））
Inspired by recent advances in the field of expert-based approximations of Gaussian processes (GPs), we present an expert-based approach to large-scale multi-output regression using single-output GP experts. Employing a deeply structured mixture of single-output GPs encoded via a probabilistic circuit allows us to capture correlations between multiple output dimensions accurately. By recursively partitioning the covariate space and the output space, posterior inference in our model reduces to inference on single-output GP experts, which only need to be conditioned on a small subset of the obse…
Inferring gene regulatory networks is an important problem in systems biology. However, these networks can be hard to infer from experimental data because of the inherent variability in biological data as well as the large number of genes involved. We propose a fast, simple method for inferring regulatory relationships between genes from knockdown experiments in the NIH LINCS dataset by calculating posterior probabilities, incorporating prior information. We show that the method is able to find previously identified edges from TRANSFAC and JASPAR and discuss the merits and limitations of this …
Objectives: The aim of this paper is to contrast the retrospective and prospective use of experts beliefs in choosing between survival models in economic evaluations. Methods: The use of experts retrospective (posterior) beliefs is discussed. A process for prospectively quantifying prior beliefs about model parameters in five standard models is described. Statistical criterion for comparing models, and the interpretation and computation of model probabilities is discussed. A case study is provided. Results: Experts have little difficulty in expressing their posterior beliefs. Information crite…
Posterior Sampling for Competitive RL: Function Approximation and Partial Observation
Hong Kong University of Science and Technology（香港科技大学）；New York University（纽约大学）；Peking University（北京大学）；Northwestern University（西北大学）；Yale University（耶鲁大学）
This paper studies posterior contraction rates in multi-category logit models with priors incorporating group sparse structures. We consider a general class of logit models that includes the well-known multinomial logit models as a special case. Group sparsity is useful when predictor variables are naturally clustered and particularly useful for variable selection in the multinomial logit models. We provide a unified platform for posterior contraction rates of group-sparse logit models that include binary logistic regression under individual sparsity. No size restriction is directly imposed on…
Existing Bayesian models, especially nonparametric Bayesian methods, rely on specially conceived priors to incorporate domain knowledge for discovering improved latent representations. While priors can affect posterior distributions through Bayes' rule, imposing posterior regularization is arguably more direct and in some cases more natural and general. In this paper, we present regularized Bayesian inference (RegBayes), a novel computational framework that performs posterior inference with a regularization term on the desired post-data posterior distribution under an information theoretical f…
Youden's index cutoff is a classifier mapping a patient's diagnostic test outcome and available covariate information to a diagnostic category. Typically the cutoff is estimated indirectly by first modeling the conditional distributions of test outcomes given diagnosis and then choosing the optimal cutoff for the estimated distributions. Here we present a Gibbs posterior distribution for direct inference on the cutoff. Our approach makes incorporating prior information about the cutoff much easier compared to existing methods, and does so without specifying probability models for the data, whi…
Based on expert opinions, informative prior elicitation for the common Weibull lifetime distribution usually presents some difficulties since it requires to elicit a two-dimensional joint prior. We consider here a reliability framework where the available expert information states directly in terms of prior predictive values (lifetimes) and not parameter values, which are less intuitive. The novelty of our procedure is to weigh the expert information by the size m of a virtual sample yielding a similar information, the prior being seen as a reference posterior. Thus, the prior calibration by t…
Aalto University（阿尔托大学）；University of Manchester（曼彻斯特大学）
Bayesian optimization (BO) is a well-established method to optimize black-box functions whose direct evaluations are costly. In this paper, we tackle the problem of incorporating expert knowledge into BO, with the goal of further accelerating the optimization, which has received very little attention so far. We design a multi-task learning architecture for this task, with the goal of jointly eliciting the expert knowledge and minimizing the objective function. In particular, this allows for the expert knowledge to be transferred into the BO task. We introduce a specific architecture based on S…
The automation of posterior inference in Bayesian data analysis has enabled experts and nonexperts alike to use more sophisticated models, engage in faster exploratory modeling and analysis, and ensure experimental reproducibility. However, standard automated posterior inference algorithms are not tractable at the scale of massive modern datasets, and modifications to make them so are typically model-specific, require expert tuning, and can break theoretical guarantees on inferential quality. Building on the Bayesian coresets framework, this work instead takes advantage of data redundancy to s…
Recursive graphical models usually underlie the statistical modelling concerning probabilistic expert systems based on Bayesian networks. This paper defines a version of these models, denoted as recursive exponential models, which have evolved by the desire to impose sophisticated domain knowledge onto local fragments of a model. Besides the structural knowledge, as specified by a given model, the statistical modelling may also include expert opinion about the values of parameters in the model. It is shown how to translate imprecise expert knowledge into approximately conjugate prior distribut…
In this study, we introduce a new approach to statistical decision theory. Without using a loss function, we select good decision rules to choice between two hypotheses. We call them "experts". They are globally unbiased but also conditionally unbiased on a family of events. We do not try to define the best expert. We define a probability distribution on the space of "experts". The measure of evidence for a hypothesis is the inductive probability of experts that decide this hypothesis, we call this measure: a "vote". We compare this point of view with the p-values. For some family of hypothese…
In the hypothesis testing framework, p-value is often computed to determine rejection of the null hypothesis or not. On the other hand, Bayesian approaches typically compute the posterior probability of the null hypothesis to evaluate its plausibility. We revisit Lindley's paradox (Lindley, 1957) and demystify the conflicting results between Bayesian and frequentist hypothesis testing procedures by casting a two-sided hypothesis as a combination of two one-sided hypotheses along the opposite directions. This can naturally circumvent the ambiguities of assigning a point mass to the null and cho…
Generalized Power Priors for Improved Bayesian Inference with Historical Data
Research Fellow, School of Mathematics and Statistics, The University of Melbourne（墨尔本大学数学与统计学学院研究员）；Professor, School of Mathematics and Statistics, The University of Melbourne（墨尔本大学数学与统计学学院教授）
External information, such as prior information or expert opinions, can play an important role in the design, analysis and interpretation of clinical trials. However, little attention has been devoted thus far to incorporating external information in clinical trials with binary outcomes, perhaps due to the perception that binary outcomes can be treated as normally-distributed outcomes by using normal approximations. In this paper we show that these two types of clinical trials could behave differently, and that special care is needed for the analysis of clinical trials with binary outcomes. In…
Motivation: Recent advances in technology for brain imaging and high-throughput genotyping have motivated studies examining the influence of genetic variation on brain structure. Wang et al. (Bioinformatics, 2012) have developed an approach for the analysis of imaging genomic studies using penalized multi-task regression with regularization based on a novel group $l_{2,1}$-norm penalty which encourages structured sparsity at both the gene level and SNP level. While incorporating a number of useful features, the proposed method only furnishes a point estimate of the regression coefficients; tec…
In this paper we introduce a novel Bayesian approach for linking multiple social networks in order to discover the same real world person having different accounts across networks. In particular, we develop a latent model that allow us to jointly characterize the network and linkage structures relying in both relational and profile data. In contrast to other existing approaches in the machine learning literature, our Bayesian implementation naturally provides uncertainty quantification via posterior probabilities for the linkage structure itself or any function of it. Our findings clearly sugg…
German Research Center for Artificial Intelligence (DFKI)（德国人工智能研究中心）；TU Kaiserslautern（凯泽斯劳滕工业大学）；National University of Sciences and Technology (NUST)（国立科学技术大学）；National Center of Artificial Intelligence（国家人工智能中心）
The promise of ANNs to automatically discover and extract useful features/patterns from data without dwelling on domain expertise although seems highly promising but comes at the cost of high reliance on large amount of accurately labeled data, which is often hard to acquire and formulate especially in time-series domains like anomaly detection, natural disaster management, predictive maintenance and healthcare. As these networks completely rely on data and ignore a very important modality i.e. expert, they are unable to harvest any benefit from the expert knowledge, which in many cases is ver…
We propose the use of finite mixtures of continuous distributions in modelling the process by which new individuals, that arrive in groups, become part of a wildlife population. We demonstrate this approach using a data set of migrating semipalmated sandpipers (Calidris pussila) for which we extend existing stopover models to allow for individuals to have different behaviour in terms of their stopover duration at the site. We demonstrate the use of reversible jump MCMC methods to derive posterior distributions for the model parameters and the models, simultaneously. The algorithm moves between…
In this paper, we are concerned with attributing meaning to the results of a Bayesian analysis for a problem which is sufficiently complex that we are unable to assert a precise correspondence between the expert probabilistic judgements of the analyst and the particular forms chosen for the prior specification and the likelihood for the analysis. In order to do this, we propose performing a finite collection of additional Bayesian analyses under alternative collections of prior and likelihood modelling judgements that we may also view as representative of our prior knowledge and the problem st…
We propose a novel Bayesian model selection technique on linear mixed-effects models to compare multiple treatments with a control. A fully Bayesian approach is implemented to estimate the marginal inclusion probabilities that provide a direct measure of the difference between treatments and the control, along with the model-averaged posterior distributions. Default priors are proposed for model selection incorporating domain knowledge and a component-wise Gibbs sampler is developed for efficient posterior computation. We demonstrate the proposed method based on simulated data and an experimen…
Shape constrained regression analysis has applications in dose-response modeling, environmental risk assessment, disease screening and many other areas. Incorporating the shape constraints can improve estimation efficiency and avoid implausible results. We propose two novel methods focusing on Bayesian monotone curve and surface estimation using Gaussian process projections. The first projects samples from an unconstrained prior, while the second projects samples from the Gaussian process posterior. Theory is developed on continuity of the projection, posterior consistency and rates of contrac…
While vaccines are crucial to end the COVID-19 pandemic, public confidence in vaccine safety has always been vulnerable. Many statistical methods have been applied to VAERS (Vaccine Adverse Event Reporting System) database to study the safety of COVID-19 vaccines. However, all these methods ignored the adverse event (AE) ontology. AEs are naturally related; for example, events of retching, dysphagia, and reflux are all related to an abnormal digestive system. Explicitly bringing AE relationships into the model can aid in the detection of true AE signals amid the noise while reducing false posi…
We provide a geometric interpretation to Bayesian inference that allows us to introduce a natural measure of the level of agreement between priors, likelihoods, and posteriors. The starting point for the construction of our geometry is the simple observation that the marginal likelihood can be regarded as an inner product between the prior and the likelihood. A key concept in our geometry is that of compatibility, a measure which is based on the same construction principles as Pearson correlation, but which can be used to assess how much the prior agrees with the likelihood, to gauge the sensi…
