Occlusive Components Analysis

We study unsupervised learning in a probabilistic generative model for occlusion. The model uses two types of latent variables: one indicates which objects are present in the image, and the other how they are ordered in depth. This depth order then determines how the positions and appearances of the objects present, specified in the model parameters, combine to form the image. We show that the object parameters can be learnt from an unlabelled set of images in which objects occlude one another. Exact maximum-likelihood learning is intractable. However, we show that tractable approximations to Expectation Maximization (EM) can be found if the training images each contain only a small number of objects on average. In numerical experiments it is shown that these approximations recover the correct set of object parameters. Experiments on a novel version of the bars test using colored bars, and experiments on more realistic data, show that the algorithm performs well in extracting the generating causes. Experiments based on the standard bars benchmark test for object learning show that the algorithm performs well in comparison to other recent component extraction approaches. The model and the learning algorithm thus connect research on occlusion with the research field of multiple-causes component extraction methods.

Two problems with variational expectation maximisation for time-series models

Variational methods are a key component of the approximate inference and learning toolbox. These methods fill an important middle ground, retaining distributional information about uncertainty in latent variables, unlike maximum a posteriori methods (MAP), and yet generally requiring less computational time than Monte Carlo Markov Chain methods. In particular the variational Expectation Maximisation (vEM) and variational Bayes algorithms, both involving variational optimisation of a free-energy, are widely used in time-series modelling. Here, we investigate the success of vEM in simple probabilistic time-series models. First we consider the inference step of vEM, and show that a consequence of the well-known compactness property of variational inference is a failure to propagate uncertainty in time, thus limiting the usefulness of the retained distributional information. In particular, the uncertainty may appear to be smallest precisely when the approximation is poorest. Second, we consider parameter learning and analytically reveal systematic biases in the parameters found by vEM. Surprisingly, simpler variational approximations (such a mean-field) can lead to less bias than more complicated structured approximations.

No evidence for an item limit in change detection.

Change detection is a classic paradigm that has been used for decades to argue that working memory can hold no more than a fixed number of items ("item-limit models"). Recent findings force us to consider the alternative view that working memory is limited by the precision in stimulus encoding, with mean precision decreasing with increasing set size ("continuous-resource models"). Most previous studies that used the change detection paradigm have ignored effects of limited encoding precision by using highly discriminable stimuli and only large changes. We conducted two change detection experiments (orientation and color) in which change magnitudes were drawn from a wide range, including small changes. In a rigorous comparison of five models, we found no evidence of an item limit. Instead, human change detection performance was best explained by a continuous-resource model in which encoding precision is variable across items and trials even at a given set size. This model accounts for comparison errors in a principled, probabilistic manner. Our findings sharply challenge the theoretical basis for most neural studies of working memory capacity.

Kernel conditional quantile estimation via reduction revisited

Quantile regression refers to the process of estimating the quantiles of a conditional distribution and has many important applications within econometrics and data mining, among other domains. In this paper, we show how to estimate these conditional quantile functions within a Bayes risk minimization framework using a Gaussian process prior. The resulting non-parametric probabilistic model is easy to implement and allows non-crossing quantile functions to be enforced. Moreover, it can directly be used in combination with tools and extensions of standard Gaussian Processes such as principled hyperparameter estimation, sparsification, and quantile regression with input-dependent noise rates. No existing approach enjoys all of these desirable properties. Experiments on benchmark datasets show that our method is competitive with state-of-the-art approaches. © 2009 IEEE.

The vibro-acoustic analysis of built-up systems using a hybrid method with parametric and non-parametric uncertainties

An existing hybrid finite element (FE)/statistical energy analysis (SEA) approach to the analysis of the mid- and high frequency vibrations of a complex built-up system is extended here to a wider class of uncertainty modeling. In the original approach, the constituent parts of the system are considered to be either deterministic, and modeled using FE, or highly random, and modeled using SEA. A non-parametric model of randomness is employed in the SEA components, based on diffuse wave theory and the Gaussian Orthogonal Ensemble (GOE), and this enables the mean and variance of second order quantities such as vibrational energy and response cross-spectra to be predicted. In the present work the assumption that the FE components are deterministic is relaxed by the introduction of a parametric model of uncertainty in these components. The parametric uncertainty may be modeled either probabilistically, or by using a non-probabilistic approach such as interval analysis, and it is shown how these descriptions can be combined with the non-parametric uncertainty in the SEA subsystems to yield an overall assessment of the performance of the system. The method is illustrated by application to an example built-up plate system which has random properties, and benchmark comparisons are made with full Monte Carlo simulations. © 2012 Elsevier Ltd. All rights reserved.

A Hierarchical Model for Ordinal Matrix Factorization

This paper proposes a hierarchical probabilistic model for ordinal matrix factorization. Unlike previous approaches, we model the ordinal nature of the data and take a principled approach to incorporating priors for the hidden variables. Two algorithms are presented for inference, one based on Gibbs sampling and one based on variational Bayes. Importantly, these algorithms may be implemented in the factorization of very large matrices with missing entries. The model is evaluated on a collaborative filtering task, where users have rated a collection of movies and the system is asked to predict their ratings for other movies. The Netflix data set is used for evaluation, which consists of around 100 million ratings. Using root mean-squared error (RMSE) as an evaluation metric, results show that the suggested model outperforms alternative factorization techniques. Results also show how Gibbs sampling outperforms variational Bayes on this task, despite the large number of ratings and model parameters. Matlab implementations of the proposed algorithms are available from cogsys.imm.dtu.dk/ordinalmatrixfactorization.

The role of human orbitofrontal cortex in value comparison for incommensurable objects.

The human orbitofrontal cortex is strongly implicated in appetitive valuation. Whether its role extends to support comparative valuation necessary to explain probabilistic choice patterns for incommensurable goods is unknown. Using a binary choice paradigm, we derived the subjective values of different bundles of goods, under conditions of both gain and loss. We demonstrate that orbitofrontal activation reflects the difference in subjective value between available options, an effect evident across valuation for both gains and losses. In contrast, activation in dorsal striatum and supplementary motor areas reflects subjects' choice probabilities. These findings indicate that orbitofrontal cortex plays a pivotal role in valuation for incommensurable goods, a critical component process in human decision making.

Confidence in beliefs about pain predicts expectancy effects on pain perception and anticipatory processing in right anterior insula.

Psychological factors play a major role in exacerbating chronic pain. Effective self-management of pain is often hindered by inaccurate beliefs about the nature of pain which lead to a high degree of emotional reactivity. Probabilistic models of perception state that greater confidence (certainty) in beliefs increases their influence on perception and behavior. In this study, we treat confidence as a metacognitive process dissociable from the content of belief. We hypothesized that confidence is associated with anticipatory activation of areas of the pain matrix involved with top-down modulation of pain. Healthy volunteers rated their beliefs about the emotional distress that experimental pain would cause, and separately rated their level of confidence in this belief. Confidence predicted the influence of anticipation cues on experienced pain. We measured brain activity during anticipation of pain using high-density EEG and used electromagnetic tomography to determine neural substrates of this effect. Confidence correlated with activity in right anterior insula, posterior midcingulate and inferior parietal cortices during the anticipation of pain. Activity in the right anterior insula predicted a greater influence of anticipation cues on pain perception, whereas activity in right inferior parietal cortex predicted a decreased influence of anticipatory cues. The results support probabilistic models of pain perception and suggest that confidence in beliefs is an important determinant of expectancy effects on pain perception.

Simulating early adoption of alternative fuel vehicles for sustainability

We quantify the conditions that might trigger wide spread adoption of alternative fuel vehicles (AFVs) to support energy policy. Empirical review shows that early adopters are heterogeneous motivated by financial benefits, environmental appeal, new technology, and vehicle reliability. A probabilistic Monte Carlo simulation model is used to assess consumer heterogeneity for early and mass market adopters. For early adopters full battery electric vehicles (BEVs) are competitive but unable to surpass diesels or hybrids due to purchase price premium and lack of charging availability. For mass adoption, simulations indicate that if the purchase price premium of a BEV closes to within 20% of an in-class internal combustion engine (ICE) vehicle, combined with a 60% increase in refuelling availability relative to the incumbent system, BEVs become competitive. But this depends on a mass market that values the fuel economy and CO2 reduction benefits associated with BEVs. We also find that the largest influence on early adoption is financial benefit rather than pro-environmental behaviour suggesting that AFVs should be marketed by appealing to economic benefits combined with pro-environmental behaviour to motivate adoption. Monte Carlo simulations combined with scenarios can give insight into diffusion dynamics for other energy demand-side technologies. © 2012 Elsevier Inc.

Active learning of model evidence using Bayesian quadrature

Numerical integration is a key component of many problems in scientific computing, statistical modelling, and machine learning. Bayesian Quadrature is a modelbased method for numerical integration which, relative to standard Monte Carlo methods, offers increased sample efficiency and a more robust estimate of the uncertainty in the estimated integral. We propose a novel Bayesian Quadrature approach for numerical integration when the integrand is non-negative, such as the case of computing the marginal likelihood, predictive distribution, or normalising constant of a probabilistic model. Our approach approximately marginalises the quadrature model's hyperparameters in closed form, and introduces an active learning scheme to optimally select function evaluations, as opposed to using Monte Carlo samples. We demonstrate our method on both a number of synthetic benchmarks and a real scientific problem from astronomy.

Reliability analysis of deep excavation based on a semi-empirical approach

The design and construction of deep excavations in urban environment is often governed by serviceability limit state related to the risk of damage to adjacent buildings. In current practice, the assessment of excavation-induced building damage has focused on a deterministic approach. This paper presents a component/system reliability analysis framework to assess the probability that specified threshold design criteria for multiple serviceability limit states are exceeded. A recently developed Bayesian probabilistic framework is used to update the predictions of ground movements in the later stages of excavation based on the recorded deformation measurements. An example is presented to show how the serviceability performance for excavation problems can be assessed based on the component/system reliability analysis. © 2011 ASCE.

Estimation of soil properties and deformations in staged constructions based on mcmc method

This paper presents a Bayesian probabilistic framework to assess soil properties and model uncertainty to better predict excavation-induced deformations using field deformation data. The potential correlations between deformations at different depths are accounted for in the likelihood function needed in the Bayesian approach. The proposed approach also accounts for inclinometer measurement errors. The posterior statistics of the unknown soil properties and the model parameters are computed using the Delayed Rejection (DR) method and the Adaptive Metropolis (AM) method. As an application, the proposed framework is used to assess the unknown soil properties of multiple soil layers using deformation data at different locations and for incremental excavation stages. The developed approach can be used for the design of optimal revisions for supported excavation systems. © 2010 ASCE.

Bayesian updating of a unified soil compression model

Some amount of differential settlement occurs even in the most uniform soil deposit, but it is extremely difficult to estimate because of the natural heterogeneity of the soil. The compression response of the soil and its variability must be characterised in order to estimate the probability of the differential settlement exceeding a certain threshold value. The work presented in this paper introduces a probabilistic framework to address this issue in a rigorous manner, while preserving the format of a typical geotechnical settlement analysis. In order to avoid dealing with different approaches for each category of soil, a simplified unified compression model is used to characterise the nonlinear compression behavior of soils of varying gradation through a single constitutive law. The Bayesian updating rule is used to incorporate information from three different laboratory datasets in the computation of the statistics (estimates of the means and covariance matrix) of the compression model parameters, as well as of the uncertainty inherent in the model.

The Supervised IBP: Neighbourhood Preserving Infinite Latent Feature Models

We propose a probabilistic model to infer supervised latent variables in the Hamming space from observed data. Our model allows simultaneous inference of the number of binary latent variables, and their values. The latent variables preserve neighbourhood structure of the data in a sense that objects in the same semantic concept have similar latent values, and objects in different concepts have dissimilar latent values. We formulate the supervised infinite latent variable problem based on an intuitive principle of pulling objects together if they are of the same type, and pushing them apart if they are not. We then combine this principle with a flexible Indian Buffet Process prior on the latent variables. We show that the inferred supervised latent variables can be directly used to perform a nearest neighbour search for the purpose of retrieval. We introduce a new application of dynamically extending hash codes, and show how to effectively couple the structure of the hash codes with continuously growing structure of the neighbourhood preserving infinite latent feature space.

Optimal recall from bounded metaplastic synapses: predicting functional adaptations in hippocampal area CA3.

A venerable history of classical work on autoassociative memory has significantly shaped our understanding of several features of the hippocampus, and most prominently of its CA3 area, in relation to memory storage and retrieval. However, existing theories of hippocampal memory processing ignore a key biological constraint affecting memory storage in neural circuits: the bounded dynamical range of synapses. Recent treatments based on the notion of metaplasticity provide a powerful model for individual bounded synapses; however, their implications for the ability of the hippocampus to retrieve memories well and the dynamics of neurons associated with that retrieval are both unknown. Here, we develop a theoretical framework for memory storage and recall with bounded synapses. We formulate the recall of a previously stored pattern from a noisy recall cue and limited-capacity (and therefore lossy) synapses as a probabilistic inference problem, and derive neural dynamics that implement approximate inference algorithms to solve this problem efficiently. In particular, for binary synapses with metaplastic states, we demonstrate for the first time that memories can be efficiently read out with biologically plausible network dynamics that are completely constrained by the synaptic plasticity rule, and the statistics of the stored patterns and of the recall cue. Our theory organises into a coherent framework a wide range of existing data about the regulation of excitability, feedback inhibition, and network oscillations in area CA3, and makes novel and directly testable predictions that can guide future experiments.