Search

The increasing availability of graph-structured data motivates the task of optimising over functions defined on the node set of graphs. Traditional graph search algorithms can be applied in this case, but they may be sample-inefficient and do not make use of information about the function values; on the other hand, Bayesian optimisation is a class of promising black-box solvers with superior sample efficiency, but it has been scarcely been applied to such novel setups. To fill this gap, we propose a novel Bayesian optimisation framework that optimises over functions defined on generic, large-scale and potentially unknown graphs. Through the learning of suitable kernels on graphs, our framework has the advantage of adapting to the behaviour of the target function. The local modelling approach further guarantees the efficiency of our method. Extensive experiments on both synthetic and real-world graphs demonstrate the effectiveness of the proposed optimisation framework., Comment: 10 pages, 9 figures, 1 table (23 pages, 24 figures, 1 table including references and appendices)

Peer reviewed: True, Funder: Structural Biology Platform of the Université de Montréal; Grant(s): RRID:SCR_022303, The eukaryotic DNA replication fork is a hub of enzymes that continuously act to synthesize DNA, propagate DNA methylation and other epigenetic marks, perform quality control, repair nascent DNA, and package this DNA into chromatin. Many of the enzymes involved in these spatiotemporally correlated processes perform their functions by binding to proliferating cell nuclear antigen (PCNA). A long-standing question has been how the plethora of PCNA-binding enzymes exert their activities without interfering with each other. As a first step towards deciphering this complex regulation, we studied how Chromatin Assembly Factor 1 (CAF-1) binds to PCNA. We demonstrate that CAF-1 binds to PCNA in a heretofore uncharacterized manner that depends upon a cation-pi (π) interaction. An arginine residue, conserved among CAF-1 homologs but absent from other PCNA-binding proteins, inserts into the hydrophobic pocket normally occupied by proteins that contain canonical PCNA interaction peptides (PIPs). Mutation of this arginine disrupts the ability of CAF-1 to bind PCNA and to assemble chromatin. The PIP of the CAF-1 p150 subunit resides at the extreme C-terminus of an apparent long α-helix (119 amino acids) that has been reported to bind DNA. The length of that helix and the presence of a PIP at the C-terminus are evolutionarily conserved among numerous species, ranging from yeast to humans. This arrangement of a very long DNA-binding coiled-coil that terminates in PIPs may serve to coordinate DNA and PCNA binding by CAF-1.

Bayesian optimization (BO) is a sample-efficient approach for tuning design parameters to optimize expensive-to-evaluate, black-box performance metrics. In many manufacturing processes, the design parameters are subject to random input noise, resulting in a product that is often less performant than expected. Although BO methods have been proposed for optimizing a single objective under input noise, no existing method addresses the practical scenario where there are multiple objectives that are sensitive to input perturbations. In this work, we propose the first multi-objective BO method that is robust to input noise. We formalize our goal as optimizing the multivariate value-at-risk (MVaR), a risk measure of the uncertain objectives. Since directly optimizing MVaR is computationally infeasible in many settings, we propose a scalable, theoretically-grounded approach for optimizing MVaR using random scalarizations. Empirically, we find that our approach significantly outperforms alternative methods and efficiently identifies optimal robust designs that will satisfy specifications across multiple metrics with high probability., Comment: To appear at ICML 2022. 36 pages. Code is available at https://github.com/facebookresearch/robust_mobo

A wide variety of battery models are available, and it is not always obvious which model `best' describes a dataset. This paper presents a Bayesian model selection approach using Bayesian quadrature. The model evidence is adopted as the selection metric, choosing the simplest model that describes the data, in the spirit of Occam's razor. However, estimating this requires integral computations over parameter space, which is usually prohibitively expensive. Bayesian quadrature offers sample-efficient integration via model-based inference that minimises the number of battery model evaluations. The posterior distribution of model parameters can also be inferred as a byproduct without further computation. Here, the simplest lithium-ion battery models, equivalent circuit models, were used to analyse the sensitivity of the selection criterion to given different datasets and model configurations. We show that popular model selection criteria, such as root-mean-square error and Bayesian information criterion, can fail to select a parsimonious model in the case of a multimodal posterior. The model evidence can spot the optimal model in such cases, simultaneously providing the variance of the evidence inference itself as an indication of confidence. We also show that Bayesian quadrature can compute the evidence faster than popular Monte Carlo based solvers., Comment: 11 pages, 2 figures, accepted at IFAC2023

Gaussian processes (GPs) produce good probabilistic models of functions, but most GP kernels require $O((n+m)n^2)$ time, where $n$ is the number of data points and $m$ the number of predictive locations. We present a new kernel that allows for Gaussian process regression in $O((n+m)\log(n+m))$ time. Our "binary tree" kernel places all data points on the leaves of a binary tree, with the kernel depending only on the depth of the deepest common ancestor. We can store the resulting kernel matrix in $O(n)$ space in $O(n \log n)$ time, as a sum of sparse rank-one matrices, and approximately invert the kernel matrix in $O(n)$ time. Sparse GP methods also offer linear run time, but they predict less well than higher dimensional kernels. On a classic suite of regression tasks, we compare our kernel against Mat\'ern, sparse, and sparse variational kernels. The binary tree GP assigns the highest likelihood to the test data on a plurality of datasets, usually achieves lower mean squared error than the sparse methods, and often ties or beats the Mat\'ern GP. On large datasets, the binary tree GP is fastest, and much faster than a Mat\'ern GP., Comment: NeurIPS 2022; 9 pages + appendices

Modern approximations to Gaussian processes are suitable for "tall data", with a cost that scales well in the number of observations, but under-performs on ``wide data'', scaling poorly in the number of input features. That is, as the number of input features grows, good predictive performance requires the number of summarising variables, and their associated cost, to grow rapidly. We introduce a kernel that allows the number of summarising variables to grow exponentially with the number of input features, but requires only linear cost in both number of observations and input features. This scaling is achieved through our introduction of the Bézier buttress, which allows approximate inference without computing matrix inverses or determinants. We show that our kernel has close similarities to some of the most used kernels in Gaussian process regression, and empirically demonstrate the kernel's ability to scale to both tall and wide datasets.

Reinforcement learning (RL) offers the potential for training generally capable agents that can interact autonomously in the real world. However, one key limitation is the brittleness of RL algorithms to core hyperparameters and network architecture choice. Furthermore, non-stationarities such as evolving training data and increased agent complexity mean that different hyperparameters and architectures may be optimal at different points of training. This motivates AutoRL, a class of methods seeking to automate these design choices. One prominent class of AutoRL methods is Population-Based Training (PBT), which have led to impressive performance in several large scale settings. In this paper, we introduce two new innovations in PBT-style methods. First, we employ trust-region based Bayesian Optimization, enabling full coverage of the high-dimensional mixed hyperparameter search space. Second, we show that using a generational approach, we can also learn both architectures and hyperparameters jointly on-the-fly in a single training run. Leveraging the new highly parallelizable Brax physics engine, we show that these innovations lead to large performance gains, significantly outperforming the tuned baseline while learning entire configurations on the fly. Code is available at https://github.com/xingchenwan/bgpbt., Comment: AutoML Conference 2022. 10 pages, 4 figure, 3 tables (28 pages, 10 figures, 7 tables including references and appendices)

KL-regularized reinforcement learning from expert demonstrations has proved successful in improving the sample efficiency of deep reinforcement learning algorithms, allowing them to be applied to challenging physical real-world tasks. However, we show that KL-regularized reinforcement learning with behavioral reference policies derived from expert demonstrations can suffer from pathological training dynamics that can lead to slow, unstable, and suboptimal online learning. We show empirically that the pathology occurs for commonly chosen behavioral policy classes and demonstrate its impact on sample efficiency and online policy performance. Finally, we show that the pathology can be remedied by non-parametric behavioral reference policies and that this allows KL-regularized reinforcement learning to significantly outperform state-of-the-art approaches on a variety of challenging locomotion and dexterous hand manipulation tasks., Comment: Published in Advances in Neural Information Processing Systems 34 (NeurIPS 2021)

Despite UNESCO’s Learning Cities agenda, which argues for the mobilisation of resources to promote education across all sectors and environments, there is little evaluative research on learning city engagement which is both naturalistic and empirically rigorous. The research on informal adult learning in urban contexts is particularly sparse. This paper provides a case study of informal learning and lifewide literacies amongst Glaswegian adults using three distinct approaches to data collection: a household survey capturing rich data on learning attitudes, behaviours, and literacies; GPS trails that track mobility around the city; and the capture of naturally occurring social media. The work operationalises learning city indicators, and explores domains beyond education, some of which have not previously been considered in surveys of adult learning, for example, physical mobilities and transportation patterns. We use the theoretical concepts of Social Identity and Social Capital to situate social inclusion within explanatory frameworks which interpret marginalisation in groups, places and the less tangible domain of informal learning in order to interpret our multi-strand dataset. A triangulated analysis of city-wide participation in lifewide learning reveals a demographic picture of groups marginalised from learning opportunities and practices. Examples of this include older adults in areas of deprivation and households in precarity. We conclude with a call for new approaches to exploring learning participation which offer novel methods to evidence informal learning and lifewide literacies.

BackgroundWhile growing evidence suggests a link between periodontaldisease (PD) and cardiovascular disease (CVD), the independence of this association and the pathway remain unclear. Herein, we tested the hypotheses that: (1) inflammation of the periodontium (PDinflammation ) predicts future CVD independently of disease risk factors shared between CVD and PD, and (2) the mechanism linking the two diseases involves heightened arterial inflammation.Methods18 F-fluorodeoxyglucose positron emission tomography/computed tomography (18 F-FDG-PET/CT) imaging was performed in 304 individuals (median age 54 years; 42.4% male) largely for cancer screening; individuals without active cancer were included. PDinflammation and arterial inflammation were quantified using validated 18 F-FDG-PET/CT methods. Additionally, we evaluated the relationship between PDinflammation and subsequent major adverse cardiovascular events (MACE) using Cox models and log-rank tests.ResultsThirteen individuals developed MACE during follow-up (median 4.1 years). PDinflammation associated with arterial inflammation, remaining significant after adjusting for PD and CVD risk factors (standardized β [95% CI]: 0.30 [0.20-0.40], P&nbsp

Many functions have approximately-known upper and/or lower bounds, potentially aiding the modeling of such functions. In this paper, we introduce Gaussian process models for functions where such bounds are (approximately) known. More specifically, we propose the first use of such bounds to improve Gaussian process (GP) posterior sampling and Bayesian optimization (BO). That is, we transform a GP model satisfying the given bounds, and then sample and weight functions from its posterior. To further exploit these bounds in BO settings, we present bounded entropy search (BES) to select the point gaining the most information about the underlying function, estimated by the GP samples, while satisfying the output constraints. We characterize the sample variance bounds and show that the decision made by BES is explainable. Our proposed approach is conceptually straightforward and can be used as a plug in extension to existing methods for GP posterior sampling and Bayesian optimization., Comment: 20 pages

Offline reinforcement learning enables agents to leverage large pre-collected datasets of environment transitions to learn control policies, circumventing the need for potentially expensive or unsafe online data collection. Significant progress has been made recently in offline model-based reinforcement learning, approaches which leverage a learned dynamics model. This typically involves constructing a probabilistic model, and using the model uncertainty to penalize rewards where there is insufficient data, solving for a pessimistic MDP that lower bounds the true MDP. Existing methods, however, exhibit a breakdown between theory and practice, whereby pessimistic return ought to be bounded by the total variation distance of the model from the true dynamics, but is instead implemented through a penalty based on estimated model uncertainty. This has spawned a variety of uncertainty heuristics, with little to no comparison between differing approaches. In this paper, we compare these heuristics, and design novel protocols to investigate their interaction with other hyperparameters, such as the number of models, or imaginary rollout horizon. Using these insights, we show that selecting these key hyperparameters using Bayesian Optimization produces superior configurations that are vastly different to those currently used in existing hand-tuned state-of-the-art methods, and result in drastically stronger performance., Comment: Spotlight @ ICLR 2022; Spotlight @ RL4RealLife Workshop ICML2021

Graph neural networks, a popular class of models effective in a wide range of graph-based learning tasks, have been shown to be vulnerable to adversarial attacks. While the majority of the literature focuses on such vulnerability in node-level classification tasks, little effort has been dedicated to analysing adversarial attacks on graph-level classification, an important problem with numerous real-life applications such as biochemistry and social network analysis. The few existing methods often require unrealistic setups, such as access to internal information of the victim models, or an impractically-large number of queries. We present a novel Bayesian optimisation-based attack method for graph classification models. Our method is black-box, query-efficient and parsimonious with respect to the perturbation applied. We empirically validate the effectiveness and flexibility of the proposed method on a wide range of graph classification tasks involving varying graph properties, constraints and modes of attack. Finally, we analyse common interpretable patterns behind the adversarial samples produced, which may shed further light on the adversarial robustness of graph classification models., Comment: NeurIPS 2021. 11 pages, 8 figures, 2 tables (24 pages, 17 figures, 8 tables including references and appendices)

High-dimensional black-box optimisation remains an important yet notoriously challenging problem. Despite the success of Bayesian optimisation methods on continuous domains, domains that are categorical, or that mix continuous and categorical variables, remain challenging. We propose a novel solution -- we combine local optimisation with a tailored kernel design, effectively handling high-dimensional categorical and mixed search spaces, whilst retaining sample efficiency. We further derive convergence guarantee for the proposed approach. Finally, we demonstrate empirically that our method outperforms the current baselines on a variety of synthetic and real-world tasks in terms of performance, computational costs, or both., Comment: ICML 2021. 9 page, 6 figures (26 pages, 16 figures, 2 tables including references and appendices)

Achieving the full promise of the Thermodynamic Variational Objective (TVO), a recently proposed variational lower bound on the log evidence involving a one-dimensional Riemann integral approximation, requires choosing a "schedule" of sorted discretization points. This paper introduces a bespoke Gaussian process bandit optimization method for automatically choosing these points. Our approach not only automates their one-time selection, but also dynamically adapts their positions over the course of optimization, leading to improved model learning and inference. We provide theoretical guarantees that our bandit optimization converges to the regret-minimizing choice of integration points. Empirical validation of our algorithm is provided in terms of improved learning and inference in Variational Autoencoders and Sigmoid Belief Networks., Comment: NeurIPS 2020

Current neural architecture search (NAS) strategies focus only on finding a single, good, architecture. They offer little insight into why a specific network is performing well, or how we should modify the architecture if we want further improvements. We propose a Bayesian optimisation (BO) approach for NAS that combines the Weisfeiler-Lehman graph kernel with a Gaussian process surrogate. Our method optimises the architecture in a highly data-efficient manner: it is capable of capturing the topological structures of the architectures and is scalable to large graphs, thus making the high-dimensional and graph-like search spaces amenable to BO. More importantly, our method affords interpretability by discovering useful network features and their corresponding impact on the network performance. Indeed, we demonstrate empirically that our surrogate model is capable of identifying useful motifs which can guide the generation of new architectures. We finally show that our method outperforms existing NAS approaches to achieve the state of the art on both closed- and open-domain search spaces., Comment: ICLR 2021. 9 pages, 5 figures, 1 table (23 pages, 14 figures and 3 tables including references and appendices)

Neural architecture search (NAS) automates the design of deep neural networks. One of the main challenges in searching complex and non-continuous architectures is to compare the similarity of networks that the conventional Euclidean metric may fail to capture. Optimal transport (OT) is resilient to such complex structure by considering the minimal cost for transporting a network into another. However, the OT is generally not negative definite which may limit its ability to build the positive-definite kernels required in many kernel-dependent frameworks. Building upon tree-Wasserstein (TW), which is a negative definite variant of OT, we develop a novel discrepancy for neural architectures, and demonstrate it within a Gaussian process surrogate model for the sequential NAS settings. Furthermore, we derive a novel parallel NAS, using quality k-determinantal point process on the GP posterior, to select diverse and high-performing architectures from a discrete set of candidates. Empirically, we demonstrate that our TW-based approaches outperform other baselines in both sequential and parallel NAS., Comment: 23 pages, camera ready ICML2021

In the Global North and increasingly in the Global South, smart city technologies are enthusiastically seen as a solution to urban problems and as an alternative to austerity. However, to move beyond a narrow technological focus, it is necessary to explore the degree to which smart initiatives are committed to building socially inclusive innovation with learning at its core. Using the particular case of the Future City Demonstrator Initiative in Glasgow, United Kingdom, the most high-profile initiative of its kind funded by government, the authors of this article assess the extent to which this smart city adopts such an inclusive approach. They use the quadruple helix model (government – academia – industry – civil society) as a starting point and develop an analytic framework composed of four strands: (1) supporting participation of citizens in decision-making; (2) implementing technological innovation which positions citizens as active users; (3) implementing technological innovation to benefit the community; and (4) evaluating technological innovation in the light of the experiences and needs of citizens. Unlike most analyses, the principal focus of this article is on the fourth element of the helix, civil society. The authors argue that Glasgow’s rhetoric of smart urbanism, while aspiring to problem-solving, devalues certain principles of human agency. They emphasise that urban change, including the city’s desire to become technologically innovative, would more fully facilitate active citizenship, social inclusion and learning opportunities for all if it were underpinned by the broader conceptions and frameworks of learning cities.

We propose Radial Bayesian Neural Networks (BNNs): a variational approximate posterior for BNNs which scales well to large models while maintaining a distribution over weight-space with full support. Other scalable Bayesian deep learning methods, like MC dropout or deep ensembles, have discrete support-they assign zero probability to almost all of the weight-space. Unlike these discrete support methods, Radial BNNs' full support makes them suitable for use as a prior for sequential inference. In addition, they solve the conceptual challenges with the a priori implausibility of weight distributions with discrete support. The Radial BNN is motivated by avoiding a sampling problem in 'mean-field' variational inference (MFVI) caused by the so-called 'soap-bubble' pathology of multivariate Gaussians. We show that, unlike MFVI, Radial BNNs are robust to hyperparameters and can be efficiently applied to a challenging real-world medical application without needing ad-hoc tweaks and intensive tuning. In fact, in this setting Radial BNNs out-perform discrete-support methods like MC dropout. Lastly, by using Radial BNNs as a theoretically principled, robust alternative to MFVI we make significant strides in a Bayesian continual learning evaluation.

Proceedings of Machine Learning Research, 97, ISSN:2640-3498, Proceedings of the 36th International Conference on Machine Learning (ICML 2019)

Efficient optimisation of black-box problems that comprise both continuous and categorical inputs is important, yet poses significant challenges. We propose a new approach, Continuous and Categorical Bayesian Optimisation (CoCaBO), which combines the strengths of multi-armed bandits and Bayesian optimisation to select values for both categorical and continuous inputs. We model this mixed-type space using a Gaussian Process kernel, designed to allow sharing of information across multiple categorical variables, each with multiple possible values; this allows CoCaBO to leverage all available data efficiently. We extend our method to the batch setting and propose an efficient selection procedure that dynamically balances exploration and exploitation whilst encouraging batch diversity. We demonstrate empirically that our method outperforms existing approaches on both synthetic and real-world optimisation tasks with continuous and categorical inputs., Comment: 16 pages

Even though the United States (U.S.) spends, on average, more money per student than most Organisation for Economic Co-operation and Development (OECD) countries, it continues to lag behind its international peers in mathematics achievement. This study, which responded to the call for educational reforms that improve the mathematics achievement of U.S. students, aimed to examine the issue of student help-seeking behaviors and teacher instructional practices as they interact to affect student mathematics achievement. The Programme for International Student Assessment (PISA) defines student help-seeking behaviors as the ways in which students have a propensity to depend on the knowledge and intellect of others, including both their peers and teachers, when attempting to solve problems. Because mathematics is perhaps the most difficult school subject, student help-seeking behaviors should be a critical component of mathematics learning and teaching. Unfortunately, the research literature is barren concerning this important educational issue. This study attempted to produce the first wave of empirical evidence and open up an avenue for future research in this less-charted academic field, with the ultimate goal being to use students’ help-seeking behaviors to improve their mathematics achievement. Using the U.S. sample of 15-year-old students from PISA 2012 (the most recent PISA assessment in which the main area of focus was mathematical literacy), this study intended to determine whether students’ help-seeking behaviors play a significant role in their mathematics achievement, whether this relationship varies from school to school, and whether teacher instructional practices contribute to the school-level variation. Due to the multilevel structure of the data, with students being nested within schools, a two-level hierarchical linear model (HLM) was employed in the analysis of the data. Multiple measures of mathematics achievement were used as the dependent variables for separate analyses. Student help-seeking behavior was used as the key student-level independent variable, while three teacher instructional practices were used as the key school-level independent variables. In addition, several student and school background characteristics were used as control variables. The findings from this study indicate that student help-seeking behavior has a statistically significant effect on all measures of student mathematics achievement, even after controlling for various student background characteristics. On the other hand, the study did not find statistically significant evidence that the effects of student help-seeking behavior on any measure of student mathematics achievement vary from school to school. Overall, the issue of student help-seeking behaviors should be considered a worthy topic to pursue in future educational research. From a practical standpoint, since students’ mathematics achievement is positively associated with their help-seeking behaviors, efforts should be made to educate mathematics teachers on how to encourage their students to be more proactive in seeking help in the learning of mathematics. Copyright: © Michael C. Osborne 2019

Graph spectral techniques for measuring graph similarity, or for learning the cluster number, require kernel smoothing. The choice of kernel function and bandwidth are typically chosen in an ad-hoc manner and heavily affect the resulting output. We prove that kernel smoothing biases the moments of the spectral density. We propose an information theoretically optimal approach to learn a smooth graph spectral density, which fully respects the moment information. Our method's computational cost is linear in the number of edges, and hence can be applied to large networks, with millions of nodes. We apply our method to the problems to graph similarity and cluster number learning, where we outperform comparable iterative spectral approaches on synthetic and real graphs., Comment: 12 pages. 9 Figures

No abstract available.

We present the first treatment of the arc length of the Gaussian Process (GP) with more than a single output dimension. GPs are commonly used for tasks such as trajectory modelling, where path length is a crucial quantity of interest. Previously, only paths in one dimension have been considered, with no theoretical consideration of higher dimensional problems. We fill the gap in the existing literature by deriving the moments of the arc length for a stationary GP with multiple output dimensions. A new method is used to derive the mean of a one-dimensional GP over a finite interval, by considering the distribution of the arc length integrand. This technique is used to derive an approximate distribution over the arc length of a vector valued GP in $\mathbb{R}^n$ by moment matching the distribution. Numerical simulations confirm our theoretical derivations., Comment: 10 pages, 4 figures, Accepted to The 20th International Conference on Artificial Intelligence and Statistics (AISTATS)

In this paper we use Gaussian Process (GP) regression to propose a novel approach for predicting volatility of financial returns by forecasting the envelopes of the time series. We provide a direct comparison of their performance to traditional approaches such as GARCH. We compare the forecasting power of three approaches: GP regression on the absolute and squared returns; regression on the envelope of the returns and the absolute returns; and regression on the envelope of the negative and positive returns separately. We use a maximum a posteriori estimate with a Gaussian prior to determine our hyperparameters. We also test the effect of hyperparameter updating at each forecasting step. We use our approaches to forecast out-of-sample volatility of four currency pairs over a 2 year period, at half-hourly intervals. From three kernels, we select the kernel giving the best performance for our data. We use two published accuracy measures and four statistical loss functions to evaluate the forecasting ability of GARCH vs GPs. In mean squared error the GP's perform 20% better than a random walk model, and 50% better than GARCH for the same data., 16 pages, 8 figures, 6 tables

Accurate on-board capacity estimation is of critical importance in lithium-ion battery applications. Battery charging/discharging often occurs under a constant current load, and hence voltage vs. time measurements under this condition may be accessible in practice. This paper presents a data-driven diagnostic technique, Gaussian Process regression for In-situ Capacity Estimation (GP-ICE), which estimates battery capacity using voltage measurements over short periods of galvanostatic operation. Unlike previous works, GP-ICE does not rely on interpreting the voltage-time data as Incremental Capacity (IC) or Differential Voltage (DV) curves. This overcomes the need to differentiate the voltage-time data (a process which amplifies measurement noise), and the requirement that the range of voltage measurements encompasses the peaks in the IC/DV curves. GP-ICE is applied to two datasets, consisting of 8 and 20 cells respectively. In each case, within certain voltage ranges, as little as 10 seconds of galvanostatic operation enables capacity estimates with approximately 2-3% RMSE., Comment: 12 pages, 10 figures, submitted to IEEE Transactions on Industrial Informatics

We propose practical extensions to Bayesian optimization for solving dynamic problems. We model dynamic objective functions using spatiotemporal Gaussian process priors which capture all the instances of the functions over time. Our extensions to Bayesian optimization use the information learnt from this model to guide the tracking of a temporally evolving minimum. By exploiting temporal correlations, the proposed method also determines when to make evaluations, how fast to make those evaluations, and it induces an appropriate budget of steps based on the available information. Lastly, we evaluate our technique on synthetic and real-world problems., Comment: 10 pages, 4 figures

We develop the first Bayesian Optimization algorithm, BLOSSOM, which selects between multiple alternative acquisition functions and traditional local optimization at each step. This is combined with a novel stopping condition based on expected regret. This pairing allows us to obtain the best characteristics of both local and Bayesian optimization, making efficient use of function evaluations while yielding superior convergence to the global minimum on a selection of optimization problems, and also halting optimization once a principled and intuitive stopping condition has been fulfilled.

Integration over non-negative integrands is a central problem in machine learning (e.g. for model averaging, (hyper-)parameter marginalisation, and computing posterior predictive distributions). Bayesian Quadrature is a probabilistic numerical integration technique that performs promisingly when compared to traditional Markov Chain Monte Carlo methods. However, in contrast to easily-parallelised MCMC methods, Bayesian Quadrature methods have, thus far, been essentially serial in nature, selecting a single point to sample at each step of the algorithm. We deliver methods to select batches of points at each step, based upon those recently presented in the Batch Bayesian Optimisation literature. Such parallelisation significantly reduces computation time, especially when the integrand is expensive to sample.

Accurately predicting the future health of batteries is necessary to ensure reliable operation, minimise maintenance costs, and calculate the value of energy storage investments. The complex nature of degradation renders data-driven approaches a promising alternative to mechanistic modelling. This study predicts the changes in battery capacity over time using a Bayesian non-parametric approach based on Gaussian process regression. These changes can be integrated against an arbitrary input sequence to predict capacity fade in a variety of usage scenarios, forming a generalised health model. The approach naturally incorporates varying current, voltage and temperature inputs, crucial for enabling real world application. A key innovation is the feature selection step, where arbitrary length current, voltage and temperature measurement vectors are mapped to fixed size feature vectors, enabling them to be efficiently used as exogenous variables. The approach is demonstrated on the open-source NASA Randomised Battery Usage Dataset, with data of 26 cells aged under randomized operational conditions. Using half of the cells for training, and half for validation, the method is shown to accurately predict non-linear capacity fade, with a best case normalised root mean square error of 4.3%, including accurate estimation of prediction uncertainty.

Supplemental Digital Content is available in the text., Background— Combined positron emission tomography (PET) and computed tomography (CT) can assess both anatomy and biology of carotid atherosclerosis. We sought to assess whether 18F-fluoride or 18F-fluorodeoxyglucose can identify culprit and high-risk carotid plaque. Methods and Results— We performed 18F-fluoride and 18F-fluorodeoxyglucose PET/CT in 26 patients after recent transient ischemic attack or minor ischemic stroke: 18 patients with culprit carotid stenosis awaiting carotid endarterectomy and 8 controls without culprit carotid atheroma. We compared standardized uptake values in the clinically adjudicated culprit to the contralateral asymptomatic artery, and assessed the relationship between radiotracer uptake and plaque phenotype or predicted cardiovascular risk (ASSIGN score [Assessing Cardiovascular Risk Using SIGN Guidelines to Assign Preventive Treatment]). We also performed micro PET/CT and histological analysis of excised plaque. On histological and micro PET/CT analysis, 18F-fluoride selectively highlighted microcalcification. Carotid 18F-fluoride uptake was increased in clinically adjudicated culprit plaques compared with asymptomatic contralateral plaques (log10standardized uptake valuemean 0.29±0.10 versus 0.23±0.11, P=0.001) and compared with control patients (log10standardized uptake valuemean 0.29±0.10 versus 0.12±0.11, P=0.001). 18F-Fluoride uptake correlated with high-risk plaque features (remodeling index [r=0.53, P=0.003], plaque burden [r=0.51, P=0.004]), and predicted cardiovascular risk [r=0.65, P=0.002]). Carotid 18F-fluorodeoxyglucose uptake appeared to be increased in 7 of 16 culprit plaques, but no overall differences in uptake were observed in culprit versus contralateral plaques or control patients. However, 18F-fluorodeoxyglucose did correlate with predicted cardiovascular risk (r=0.53, P=0.019), but not with plaque phenotype. Conclusions— 18F-Fluoride PET/CT highlights culprit and phenotypically high-risk carotid plaque. This has the potential to improve risk stratification and selection of patients who may benefit from intervention.

We present the first general purpose framework for marginal maximum a posteriori estimation of probabilistic program variables. By using a series of code transformations, the evidence of any probabilistic program, and therefore of any graphical model, can be optimized with respect to an arbitrary subset of its sampled variables. To carry out this optimization, we develop the first Bayesian optimization package to directly exploit the source code of its target, leading to innovations in problem-independent hyperpriors, unbounded optimization, and implicit constraint satisfaction; delivering significant performance improvements over prominent existing packages. We present applications of our method to a number of tasks including engineering design and parameter optimization.

Information-theoretic Bayesian optimisation techniques have demonstrated state-of-the-art performance in tackling important global optimisation problems. However, current information-theoretic approaches require many approximations in implementation, introduce often-prohibitive computational overhead and limit the choice of kernels available to model the objective. We develop a fast information-theoretic Bayesian Optimisation method, FITBO, that avoids the need for sampling the global minimiser, thus significantly reducing computational overhead. Moreover, in comparison with existing approaches, our method faces fewer constraints on kernel choice and enjoys the merits of dealing with the output space. We demonstrate empirically that FITBO inherits the performance associated with information-theoretic Bayesian optimisation, while being even faster than simpler Bayesian optimisation approaches, such as Expected Improvement., Comment: Main Paper: 9 pages, 6 figures, 2 tables; Accepted by ICML 2018

The computational and storage complexity of kernel machines presents the primary barrier to their scaling to large, modern, datasets. A common way to tackle the scalability issue is to use the conjugate gradient algorithm, which relieves the constraints on both storage (the kernel matrix need not be stored) and computation (both stochastic gradients and parallelization can be used). Even so, conjugate gradient is not without its own issues: the conditioning of kernel matrices is often such that conjugate gradients will have poor convergence in practice. Preconditioning is a common approach to alleviating this issue. Here we propose preconditioned conjugate gradients for kernel machines, and develop a broad range of preconditioners particularly useful for kernel matrices. We describe a scalable approach to both solving kernel machines and learning their hyperparameters. We show this approach is exact in the limit of iterations and outperforms state-of-the-art approximations for a given computational budget.

Additional estimations. (DOCX 52 kb)

We present Blitzkriging, a new approach to fast inference for Gaussian processes, applicable to regression, optimisation and classification. State-of-the-art (stochastic) inference for Gaussian processes on very large datasets scales cubically in the number of 'inducing inputs', variables introduced to factorise the model. Blitzkriging shares state-of-the-art scaling with data, but reduces the scaling in the number of inducing points to approximately linear. Further, in contrast to other methods, Blitzkriging: does not force the data to conform to any particular structure (including grid-like); reduces reliance on error-prone optimisation of inducing point locations; and is able to learn rich (covariance) structure from the data. We demonstrate the benefits of our approach on real data in regression, time-series prediction and signal-interpolation experiments.

We present GLASSES: Global optimisation with Look-Ahead through Stochastic Simulation and Expected-loss Search. The majority of global optimisation approaches in use are myopic, in only considering the impact of the next function value; the non-myopic approaches that do exist are able to consider only a handful of future evaluations. Our novel algorithm, GLASSES, permits the consideration of dozens of evaluations into the future. This is done by approximating the ideal look-ahead loss function, which is expensive to evaluate, by a cheaper alternative in which the future steps of the algorithm are simulated beforehand. An Expectation Propagation algorithm is used to compute the expected value of the loss.We show that the far-horizon planning thus enabled leads to substantive performance gains in empirical tests., Comment: 12 pages, 9 figures

This article provides a brief overview of historic work in the field of Learning City \ud development. It then proceeds to highlight two contemporary strands of work. The first \ud is the initiative of UNESCO’s Institute for Lifelong Learning (UIL) in establishing the \ud International Platform of Learning Cities. The second is the work of the PASCAL \ud Observatory, currently manifested in the Learning Cities 2020 programme.

Massive Open Online Courses (MOOCs) bring together thousands of people from different geographies and demographic backgrounds -- but to date, little is known about how they learn or communicate. We introduce a new content-analysed MOOC dataset and use Bayesian Non-negative Matrix Factorization (BNMF) to extract communities of learners based on the nature of their online forum posts. We see that BNMF yields a superior probabilistic generative model for online discussions when compared to other models, and that the communities it learns are differentiated by their composite students' demographic and course performance indicators. These findings suggest that computationally efficient probabilistic generative modelling of MOOCs can reveal important insights for educational researchers and practitioners and help to develop more intelligent and responsive online learning environments., Comment: 10 pages, 3 figures, 1 table. Submitted for review to UAI 2014

We introduce a means of automating machine learning (ML) for big data tasks, by performing scalable stochastic Bayesian optimisation of ML algorithm parameters and hyper-parameters. More often than not, the critical tuning of ML algorithm parameters has relied on domain expertise from experts, along with laborious hand-tuning, brute search or lengthy sampling runs. Against this background, Bayesian optimisation is finding increasing use in automating parameter tuning, making ML algorithms accessible even to non-experts. However, the state of the art in Bayesian optimisation is incapable of scaling to the large number of evaluations of algorithm performance required to fit realistic models to complex, big data. We here describe a stochastic, sparse, Bayesian optimisation strategy to solve this problem, using many thousands of noisy evaluations of algorithm performance on subsets of data in order to effectively train algorithms for big data. We provide a comprehensive benchmarking of possible sparsification strategies for Bayesian optimisation, concluding that a Nystrom approximation offers the best scaling and performance for real tasks. Our proposed algorithm demonstrates substantial improvement over the state of the art in tuning the parameters of a Gaussian Process time series prediction task on real, big data.

Existing work in multi-agent collision prediction and avoidance typically assumes discrete-time trajectories with Gaussian uncertainty or that are completely deterministic. We propose an approach that allows detection of collisions even between continuous, stochastic trajectories with the only restriction that means and variances can be computed. To this end, we employ probabilistic bounds to derive criterion functions whose negative sign provably is indicative of probable collisions. For criterion functions that are Lipschitz, an algorithm is provided to rapidly find negative values or prove their absence. We propose an iterative policy-search approach that avoids prior discretisations and yields collision-free trajectories with adjustably high certainty. We test our method with both fixed-priority and auction-based protocols for coordinating the iterative planning process. Results are provided in collision-avoidance simulations of feedback controlled plants., Comment: This preprint is an extended version of a conference paper that is to appear in \textit{Proceedings of the 13th International Conference on Autonomous Agents and Multiagent Systems (AAMAS 2014)}

We propose an active learning method for discovering low-dimensional structure in high-dimensional Gaussian process (GP) tasks. Such problems are increasingly frequent and important, but have hitherto presented severe practical difficulties. We further introduce a novel technique for approximately marginalizing GP hyperparameters, yielding marginal predictions robust to hyperparameter mis-specification. Our method offers an efficient means of performing GP regression, quadrature, or Bayesian optimization in high-dimensional spaces.

We propose a novel sampling framework for inference in probabilistic models: an active learning approach that converges more quickly (in wall-clock time) than Markov chain Monte Carlo (MCMC) benchmarks. The central challenge in probabilistic inference is numerical integration, to average over ensembles of models or unknown (hyper-)parameters (for example to compute the marginal likelihood or a partition function). MCMC has provided approaches to numerical integration that deliver state-of-the-art inference, but can suffer from sample inefficiency and poor convergence diagnostics. Bayesian quadrature techniques offer a model-based solution to such problems, but their uptake has been hindered by prohibitive computation costs. We introduce a warped model for probabilistic integrands (likelihoods) that are known to be non-negative, permitting a cheap active learning scheme to optimally select sample locations. Our algorithm is demonstrated to offer faster convergence (in seconds) relative to simple Monte Carlo and annealed importance sampling on both synthetic and real-world examples.

In practical Bayesian optimization, we must often search over structures with differing numbers of parameters. For instance, we may wish to search over neural network architectures with an unknown number of layers. To relate performance data gathered for different architectures, we define a new kernel for conditional parameter spaces that explicitly includes information about which parameters are relevant in a given structure. We show that this kernel improves model quality and Bayesian optimization results over several simpler baseline kernels., Comment: 6 pages, 3 figures. Appeared in the NIPS 2013 workshop on Bayesian optimization

This paper presents a Bayesian generative model for dependent Cox point processes, alongside an efficient inference scheme which scales as if the point processes were modelled independently. We can handle missing data naturally, infer latent structure, and cope with large numbers of observed processes. A further novel contribution enables the model to work effectively in higher dimensional spaces. Using this method, we achieve vastly improved predictive performance on both 2D and 1D real data, validating our structured approach., Comment: Presented at UAI 2014. Bibtex: @inproceedings{structcoxpp14_UAI, Author = {Tom Gunter and Chris Lloyd and Michael A. Osborne and Stephen J. Roberts}, Title = {Efficient Bayesian Nonparametric Modelling of Structured Point Processes}, Booktitle = {Uncertainty in Artificial Intelligence (UAI)}, Year = {2014}}

We present the first fully variational Bayesian inference scheme for continuous Gaussian-process-modulated Poisson processes. Such point processes are used in a variety of domains, including neuroscience, geo-statistics and astronomy, but their use is hindered by the computational cost of existing inference schemes. Our scheme: requires no discretisation of the domain; scales linearly in the number of observed events; and is many orders of magnitude faster than previous sampling based approaches. The resulting algorithm is shown to outperform standard methods on synthetic examples, coal mining disaster data and in the prediction of Malaria incidences in Kenya., Comment: in ICML 2015