Probabilistic topic models such as latent Dirichlet allocation (LDA) are popularly used with Bayesian inference methods such as Gibbs sampling to learn posterior distributions over topic model parameters. We derive a novel measure of LDA topic quality using the variability of the posterior distributions. Compared to several existing baselines for automatic topic evaluation, the proposed metric achieves state-of-the-art correlations with human judgments of topic quality in experiments on three corpora. We additionally demonstrate that topic quality estimation can be further improved using a supervised estimator that combines multiple metrics.
We present a framework for Nesterov's accelerated gradient flows in probability space to design efficient mean-field Markov chain Monte Carlo (MCMC) algorithms for Bayesian inverse problems. Here four examples of information metrics are considered, including Fisher-Rao metric, Wasserstein-2 metric, Kalman-Wasserstein metric and Stein metric. For both Fisher-Rao and Wasserstein-2 metrics, we prove convergence properties of accelerated gradient flows. In implementations, we propose a sampling-efficient discrete-time algorithm for Wasserstein-2, Kalman-Wasserstein and Stein accelerated gradient flows with a restart technique. We also formulate a kernel bandwidth selection method, which learns the gradient of logarithm of density from Brownian-motion samples. Numerical experiments, including Bayesian logistic regression and Bayesian neural network, show the strength of the proposed methods compared with state-of-the-art algorithms.
While Bayesian neural networks (BNNs) hold the promise of being flexible, well-calibrated statistical models, inference often requires approximations whose consequences are poorly understood. We study the quality of common variational methods in approximating the Bayesian predictive distribution. For single-hidden layer ReLU BNNs, we prove a fundamental limitation in function-space of two of the most commonly used distributions defined in weight-space: mean-field Gaussian and Monte Carlo dropout. We find there are simple cases where neither method can have substantially increased uncertainty in between well-separated regions of low uncertainty. We provide strong empirical evidence that exact inference does not have this pathology, hence it is due to the approximation and not the model. In contrast, for deep networks, we prove a universality result showing that there exist approximate posteriors in the above classes which provide flexible uncertainty estimates. However, we find empirically that pathologies of a similar form as in the single-hidden layer case can persist when performing variational inference in deeper networks. Our results motivate careful consideration of the implications of approximate inference methods in BNNs.
A framework to boost the efficiency of Bayesian inference in probabilistic programs is introduced by embedding a sampler inside a variational posterior approximation. We call it the refined variational approximation. Its strength lies both in ease of implementation and automatically tuning of the sampler parameters to speed up mixing time using automatic differentiation. Several strategies to approximate \emph{evidence lower bound} (ELBO) computation are introduced. Experimental evidence of its efficient performance is shown solving an influence diagram in a high-dimensional space using a conditional variational autoencoder (cVAE) as a deep Bayes classifier; an unconditional VAE on density estimation tasks; and state-space models for time-series data.
As big spatial data becomes increasingly prevalent, classical spatiotemporal (ST) methods often do not scale well. While methods have been developed to account for high-dimensional spatial objects, the setting where there are exceedingly large samples of spatial observations has had less attention. The variational autoencoder (VAE), an unsupervised generative model based on deep learning and approximate Bayesian inference, fills this void using a latent variable specification that is inferred jointly across the large number of samples. In this manuscript, we compare the performance of the VAE with a more classical ST method when analyzing longitudinal visual fields from a large cohort of patients in a prospective glaucoma study. Through simulation and a case study, we demonstrate that the VAE is a scalable method for analyzing ST data, when the goal is to obtain accurate predictions. R code to implement the VAE can be found on GitHub: https://github.com/berchuck/vaeST.
There is an urgent societal need to assess whether autonomous vehicles (AVs) are safe enough. From published quantitative safety and reliability assessments of AVs, we know that, given the goal of predicting very low rates of accidents, road testing alone requires infeasible numbers of miles to be driven. However, previous analyses do not consider any knowledge prior to road testing - knowledge which could bring substantial advantages if the AV design allows strong expectations of safety before road testing. We present the advantages of a new variant of Conservative Bayesian Inference (CBI), which uses prior knowledge while avoiding optimistic biases. We then study the trend of disengagements (take-overs by human drivers) by applying Software Reliability Growth Models (SRGMs) to data from Waymo's public road testing over 51 months, in view of the practice of software updates during this testing. Our approach is to not trust any specific SRGM, but to assess forecast accuracy and then improve forecasts. We show that, coupled with accuracy assessment and recalibration techniques, SRGMs could be a valuable test planning aid.
Variational inference is becoming more and more popular for approximating intractable posterior distributions in Bayesian statistics and machine learning. Meanwhile, a few recent works have provided theoretical justification and new insights on deep neural networks for estimating smooth functions in usual settings such as nonparametric regression. In this paper, we show that variational inference for sparse deep learning retains the same generalization properties than exact Bayesian inference. In particular, we highlight the connection between estimation and approximation theories via the classical bias-variance trade-off and show that it leads to near-minimax rates of convergence for H\"older smooth functions. Additionally, we show that the model selection framework over the neural network architecture via ELBO maximization does not overfit and adaptively achieves the optimal rate of convergence.
Priors in Bayesian analyses often encode informative domain knowledge that can be useful in making the inference process more efficient. Occasionally, however, priors may be unrepresentative of the parameter values for a given dataset, which can result in inefficient parameter space exploration, or even incorrect inferences, particularly for nested sampling (NS) algorithms. Simply broadening the prior in such cases may be inappropriate or impossible in some applications. Hence our previous solution to this problem, known as posterior repartitioning (PR), redefines the prior and likelihood while keeping their product fixed, so that the posterior inferences and evidence estimates remain unchanged, but the efficiency of the NS process is significantly increased. In its most practical form, PR raises the prior to some power beta, which is introduced as an auxiliary variable that must be determined on a case-by-case basis, usually by lowering beta from unity according to some pre-defined `annealing schedule' until the resulting inferences converge to a consistent solution. Here we present a very simple yet powerful alternative Bayesian approach, in which beta is instead treated as a hyperparameter that is inferred from the data alongside the original parameters of the problem, and then marginalised over to obtain the final inference. We show through numerical examples that this Bayesian PR (BPR) method provides a very robust, self-adapting and computationally efficient `hands-off' solution to the problem of unrepresentative priors in Bayesian inference using NS. Moreover, unlike the original PR method, we show that even for representative priors BPR has a negligible computational overhead relative to standard nesting sampling, which suggests that it should be used as the default in all NS analyses.
In this paper, we describe an integrated framework for autonomous decision making in a dynamic and interactive environment. We model the interactions between the ego agent and its operating environment as a two-player dynamic game, and integrate cognitive behavioral models, Bayesian inference, and receding-horizon optimal control to define a dynamically-evolving decision strategy for the ego agent. Simulation examples representing autonomous vehicle control in three traffic scenarios where the autonomous ego vehicle interacts with a human-driven vehicle are reported.
Bayesian inference promises to ground and improve the performance of deep neural networks. It promises to be robust to overfitting, to simplify the training procedure and the space of hyperparameters, and to provide a calibrated measure of uncertainty that can enhance decision making, agent exploration and prediction fairness. Markov Chain Monte Carlo (MCMC) methods enable Bayesian inference by generating samples from the posterior distribution over model parameters. Despite the theoretical advantages of Bayesian inference and the similarity between MCMC and optimization methods, the performance of sampling methods has so far lagged behind optimization methods for large scale deep learning tasks. We aim to fill this gap and introduce ATMC, an adaptive noise MCMC algorithm that estimates and is able to sample from the posterior of a neural network. ATMC dynamically adjusts the amount of momentum and noise applied to each parameter update in order to compensate for the use of stochastic gradients. We use a ResNet architecture without batch normalization to test ATMC on the Cifar10 benchmark and the large scale ImageNet benchmark and show that, despite the absence of batch normalization, ATMC outperforms a strong optimization baseline in terms of both classification accuracy and test log-likelihood. We show that ATMC is intrinsically robust to overfitting on the training data and that ATMC provides a better calibrated measure of uncertainty compared to the optimization baseline.
Variational Bayes (VB) has become a widely-used tool for Bayesian inference in statistics and machine learning. Nonetheless, the development of the existing VB algorithms is so far generally restricted to the case where the variational parameter space is Euclidean, which hinders the potential broad application of VB methods. This paper extends the scope of VB to the case where the variational parameter space is a Riemannian manifold. We develop an efficient manifold-based VB algorithm that exploits both the geometric structure of the constraint parameter space and the information geometry of the manifold of VB approximating probability distributions. Our algorithm is provably convergent and achieves a convergence rate of order $\mathcal O(1/\sqrt{T})$ and $\mathcal O(1/T^{2-2\epsilon})$ for a non-convex evidence lower bound function and a strongly retraction-convex evidence lower bound function, respectively. We develop in particular two manifold VB algorithms, Manifold Gaussian VB and Manifold Neural Net VB, and demonstrate through numerical experiments that the proposed algorithms are stable, less sensitive to initialization and compares favourably to existing VB methods.
Inference and decision making under uncertainty are key processes in every autonomous system and numerous robotic problems. In recent years, the similarities between inference and decision making triggered much work, from developing unified computational frameworks to pondering about the duality between the two. In spite of these efforts, inference and control, as well as inference and belief space planning (BSP) are still treated as two separate processes. In this paper we propose a paradigm shift, a novel approach which deviates from conventional Bayesian inference and utilizes the similarities between inference and BSP. We make the key observation that inference can be efficiently updated using predictions made during the decision making stage, even in light of inconsistent data association between the two. We developed a two staged process that implements our novel approach and updates inference using calculations from the precursory planning phase. Using autonomous navigation in an unknown environment along with iSAM2 efficient methodologies as a test case, we benchmarked our novel approach against standard Bayesian inference, both with synthetic and real-world data (KITTI dataset). Results indicate that not only our approach improves running time by at least a factor of two while providing the same estimation accuracy, but it also alleviates the computational burden of state dimensionality and loop closures.
Deep learning (DL) has shown great potential in medical image enhancement problems, such as super-resolution or image synthesis. However, to date, little consideration has been given to uncertainty quantification over the output image. Here we introduce methods to characterise different components of uncertainty in such problems and demonstrate the ideas using diffusion MRI super-resolution. Specifically, we propose to account for $intrinsic$ uncertainty through a heteroscedastic noise model and for $parameter$ uncertainty through approximate Bayesian inference, and integrate the two to quantify $predictive$ uncertainty over the output image. Moreover, we introduce a method to propagate the predictive uncertainty on a multi-channelled image to derived scalar parameters, and separately quantify the effects of intrinsic and parameter uncertainty therein. The methods are evaluated for super-resolution of two different signal representations of diffusion MR images---DTIs and Mean Apparent Propagator MRI---and their derived quantities such as MD and FA, on multiple datasets of both healthy and pathological human brains. Results highlight three key benefits of uncertainty modelling for improving the safety of DL-based image enhancement systems. Firstly, incorporating uncertainty improves the predictive performance even when test data departs from training data. Secondly, the predictive uncertainty highly correlates with errors, and is therefore capable of detecting predictive "failures". Results demonstrate that such an uncertainty measure enables subject-specific and voxel-wise risk assessment of the output images. Thirdly, we show that the method for decomposing predictive uncertainty into its independent sources provides high-level "explanations" for the performance by quantifying how much uncertainty arises from the inherent difficulty of the task or the limited training examples.
The quantification of myocardial perfusion MRI has the potential to provide a fast, automated and user-independent assessment of myocardial ischaemia. However, due to the relatively high noise level and low temporal resolution of the acquired data and the complexity of the tracer-kinetic models, the model fitting can yield unreliable parameter estimates. A solution to this problem is the use of Bayesian inference which can incorporate prior knowledge and improve the reliability of the parameter estimation. This, however, uses Markov chain Monte Carlo sampling to approximate the posterior distribution of the kinetic parameters which is extremely time intensive. This work proposes training convolutional networks to directly predict the kinetic parameters from the signal-intensity curves that are trained using estimates obtained from the Bayesian inference. This allows fast estimation of the kinetic parameters with a similar performance to the Bayesian inference.
Bayesian inference is used extensively to infer and to quantify the uncertainty in a field of interest from a measurement of a related field when the two are linked by a physical model. Despite its many applications, Bayesian inference faces challenges when inferring fields that have discrete representations of large dimension, and/or have prior distributions that are difficult to represent mathematically. In this manuscript we consider the use of Generative Adversarial Networks (GANs) in addressing these challenges. A GAN is a type of deep neural network equipped with the ability to learn the distribution implied by multiple samples of a given field. Once trained on these samples, the generator component of a GAN maps the iid components of a low-dimensional latent vector to an approximation of the distribution of the field of interest. In this work we demonstrate how this approximate distribution may be used as a prior in a Bayesian update, and how it addresses the challenges associated with characterizing complex prior distributions and the large dimension of the inferred field. We demonstrate the efficacy of this approach by applying it to the problem of inferring and quantifying uncertainty in the initial temperature field in a heat conduction problem from a noisy measurement of the temperature at later time.
We first introduce a novel profile-based alignment algorithm, the multiple continuous Signal Alignment algorithm with Gaussian Process Regression profiles (SA-GPR). SA-GPR addresses the limitations of currently available signal alignment methods by adopting a hybrid of the particle smoothing and Markov-chain Monte Carlo (MCMC) algorithms to align signals, and by applying the Gaussian process regression to construct profiles to be aligned continuously. SA-GPR shares all the strengths of the existing alignment algorithms that depend on profiles but is more exact in the sense that profiles do not need to be discretized as sequential bins. The uncertainty of performance over the resolution of such bins is thereby eliminated. This methodology produces alignments that are consistent, that regularize extreme cases, and that properly reflect the inherent uncertainty. Then we extend SA-GPR to a specific problem in the field of paleoceanography with a method called Bayesian Inference Gaussian Process Multiproxy Alignment of Continuous Signals (BIGMACS). The goal of BIGMACS is to infer continuous ages for ocean sediment cores using two classes of age proxies: proxies that explicitly return calendar ages (e.g., radiocarbon) and those used to synchronize ages in multiple marine records (e.g., an oxygen isotope based marine proxy known as benthic ${\delta}^{18}{\rm O}$). BIGMACS integrates these two proxies by iteratively performing two steps: profile construction from benthic ${\delta}^{18}{\rm O}$ age models and alignment of each core to the profile also reflecting radiocarbon dates. We use BIGMACS to construct a new Deep Northeastern Atlantic stack (i.e., a profile from a particular benthic ${\delta}^{18}{\rm O}$ records) of five ocean sediment cores. We conclude by constructing multiproxy age models for two additional cores from the same region by aligning them to the stack.
Data-driven discovery of differential equations has been an emerging research topic. We propose a novel algorithm subsampling-based threshold sparse Bayesian regression (SubTSBR) to tackle high noise and outliers. The subsampling technique is used for improving the accuracy of the Bayesian learning algorithm. It has two parameters: subsampling size and the number of subsamples. When the subsampling size increases with fixed total sample size, the accuracy of our algorithm goes up and then down. When the number of subsamples increases, the accuracy of our algorithm keeps going up. We demonstrate how to use our algorithm step by step and compare our algorithm with threshold sparse Bayesian regression (TSBR) for the discovery of differential equations. We show that our algorithm produces better results. We also discuss the merits of discovering differential equations from data and demonstrate how to discover models with random initial and boundary condition as well as models with bifurcations. The numerical examples are: (1) predator-prey model with noise, (2) shallow water equations with outliers, (3) heat diffusion with random initial and boundary condition, and (4) fish-harvesting problem with bifurcations.
Current approaches to amortizing Bayesian inference focus solely on approximating the posterior distribution. Typically, this approximation is, in turn, used to calculate expectations for one or more target functions - a computational pipeline which is inefficient when the target function(s) are known upfront. In this paper, we address this inefficiency by introducing AMCI, a method for amortizing Monte Carlo integration directly. AMCI operates similarly to amortized inference but produces three distinct amortized proposals, each tailored to a different component of the overall expectation calculation. At runtime, samples are produced separately from each amortized proposal, before being combined to an overall estimate of the expectation. We show that while existing approaches are fundamentally limited in the level of accuracy they can achieve, AMCI can theoretically produce arbitrarily small errors for any integrable target function using only a single sample from each proposal at runtime. We further show that it is able to empirically outperform the theoretically optimal self-normalized importance sampler on a number of example problems. Furthermore, AMCI allows not only for amortizing over datasets but also amortizing over target functions.
Bayesian inference was once a gold standard for learning with neural networks, providing accurate full predictive distributions and well calibrated uncertainty. However, scaling Bayesian inference techniques to deep neural networks is challenging due to the high dimensionality of the parameter space. In this paper, we construct low-dimensional subspaces of parameter space, such as the first principal components of the stochastic gradient descent (SGD) trajectory, which contain diverse sets of high performing models. In these subspaces, we are able to apply elliptical slice sampling and variational inference, which struggle in the full parameter space. We show that Bayesian model averaging over the induced posterior in these subspaces produces accurate predictions and well calibrated predictive uncertainty for both regression and image classification.
We consider Bayesian inference of signals with vector-valued entries. Extending concentration techniques from the mathematical physics of spin glasses, we show that the matrix-valued minimum mean-square error concentrates when the size of the problem increases. Such results are often crucial for proving single-letter formulas for the mutual information when they exist. Our proof is valid in the optimal Bayesian inference setting, meaning that it relies on the assumption that the model and all its hyper-parameters are known. Examples of inference and learning problems covered by our results are spiked matrix and tensor models, the committee machine neural network with few hidden neurons in the teacher-student scenario, or multi-layers generalized linear models.