John Shawe-Taylor

Najiba Toron, Janaina Mourao-Miranda, John Shawe-Taylor

Novelty detection, a widely studied problem in machine learning, is the problem of detecting a novel class of data that has not been previously observed. A common setting for novelty detection is inductive whereby only examples of the negative class are available during training time. Transductive novelty detection on the other hand has only witnessed a recent surge in interest, it not only makes use of the negative class during training but also incorporates the (unlabeled) test set to detect novel examples. Several studies have emerged under the transductive setting umbrella that have demonstrated its advantage over its inductive counterpart. Depending on the assumptions about the data, these methods go by different names (e.g. transductive novelty detection, semi-supervised novelty detection, positive-unlabeled learning, out-of-distribution detection). With the use of generative adversarial networks (GAN), a segment of those studies have adopted a transductive setup in order to learn how to generate examples of the novel class. In this study, we propose TransductGAN, a transductive generative adversarial network that attempts to learn how to generate image examples from both the novel and negative classes by using a mixture of two Gaussians in the latent space. It achieves that by incorporating an adversarial autoencoder with a GAN network, the ability to generate examples of novel data points offers not only a visual representation of novelties, but also overcomes the hurdle faced by many inductive methods of how to tune the model hyperparameters at the decision rule level. Our model has shown superior performance over state-of-the-art inductive and transductive methods. Our study is fully reproducible with the code available publicly.

Dino Pedreschi, Luca Pappalardo, Emanuele Ferragina et al.

Human-AI coevolution, defined as a process in which humans and AI algorithms continuously influence each other, increasingly characterises our society, but is understudied in artificial intelligence and complexity science literature. Recommender systems and assistants play a prominent role in human-AI coevolution, as they permeate many facets of daily life and influence human choices on online platforms. The interaction between users and AI results in a potentially endless feedback loop, wherein users' choices generate data to train AI models, which, in turn, shape subsequent user preferences. This human-AI feedback loop has peculiar characteristics compared to traditional human-machine interaction and gives rise to complex and often ``unintended'' social outcomes. This paper introduces Coevolution AI as the cornerstone for a new field of study at the intersection between AI and complexity science focused on the theoretical, empirical, and mathematical investigation of the human-AI feedback loop. In doing so, we: (i) outline the pros and cons of existing methodologies and highlight shortcomings and potential ways for capturing feedback loop mechanisms; (ii) propose a reflection at the intersection between complexity science, AI and society; (iii) provide real-world examples for different human-AI ecosystems; and (iv) illustrate challenges to the creation of such a field of study, conceptualising them at increasing levels of abstraction, i.e., technical, epistemological, legal and socio-political.

Simon Schmitt, John Shawe-Taylor, Hado van Hasselt

How to efficiently explore in reinforcement learning is an open problem. Many exploration algorithms employ the epistemic uncertainty of their own value predictions -- for instance to compute an exploration bonus or upper confidence bound. Unfortunately the required uncertainty is difficult to estimate in general with function approximation. We propose epistemic value estimation (EVE): a recipe that is compatible with sequential decision making and with neural network function approximators. It equips agents with a tractable posterior over all their parameters from which epistemic value uncertainty can be computed efficiently. We use the recipe to derive an epistemic Q-Learning agent and observe competitive performance on a series of benchmarks. Experiments confirm that the EVE recipe facilitates efficient exploration in hard exploration tasks.

Sahan Bulathwela, Meghana Verma, Maria Perez-Ortiz et al.

This work explores how population-based engagement prediction can address cold-start at scale in large learning resource collections. The paper introduces i) VLE, a novel dataset that consists of content and video based features extracted from publicly available scientific video lectures coupled with implicit and explicit signals related to learner engagement, ii) two standard tasks related to predicting and ranking context-agnostic engagement in video lectures with preliminary baselines and iii) a set of experiments that validate the usefulness of the proposed dataset. Our experimental results indicate that the newly proposed VLE dataset leads to building context-agnostic engagement prediction models that are significantly performant than ones based on previous datasets, mainly attributing to the increase of training examples. VLE dataset's suitability in building models towards Computer Science/ Artificial Intelligence education focused on e-learning/ MOOC use-cases is also evidenced. Further experiments in combining the built model with a personalising algorithm show promising improvements in addressing the cold-start problem encountered in educational recommenders. This is the largest and most diverse publicly available dataset to our knowledge that deals with learner engagement prediction tasks. The dataset, helper tools, descriptive statistics and example code snippets are available publicly.

Janaina Mourão-Miranda, Zakria Hussain, Konstantinos Tsirlis et al.

Multiple Kernel Learning (MKL) models combine several kernels in supervised and unsupervised settings to integrate multiple data representations or sources, each represented by a different kernel. MKL seeks an optimal linear combination of base kernels that maximizes a generalized performance measure under a regularization constraint. Various norms have been used to regularize the kernel weights, including $l1$, $l2$ and $lp$, as well as the "elastic-net" penalty, which combines $l1$- and $l2$-norm to promote both sparsity and the selection of correlated kernels. This property makes elastic-net regularized MKL (ENMKL) especially valuable when model interpretability is critical and kernels capture correlated information, such as in neuroimaging. Previous ENMKL methods have followed a two-stage procedure: fix kernel weights, train a support vector machine (SVM) with the weighted kernel, and then update the weights via gradient descent, cutting-plane methods, or surrogate functions. Here, we introduce an alternative ENMKL formulation that yields a simple analytical update for the kernel weights. We derive explicit algorithms for both SVM and kernel ridge regression (KRR) under this framework, and implement them in the open-source Pattern Recognition for Neuroimaging Toolbox (PRoNTo). We evaluate these ENMKL algorithms against $l1$-norm MKL and against SVM (or KRR) trained on the unweighted sum of kernels across three neuroimaging applications. Our results show that ENMKL matches or outperforms $l1$-norm MKL in all tasks and only underperforms standard SVM in one scenario. Crucially, ENMKL produces sparser, more interpretable models by selectively weighting correlated kernels.

Diego García-Pérez, Emilio Parrado-Hernández, John Shawe-Taylor

This paper presents four theoretical contributions that improve the usability of risk certificates for neural networks based on PAC-Bayes bounds. First, two bounds on the KL divergence between Bernoulli distributions enable the derivation of the tightest explicit bounds on the true risk of classifiers across different ranges of empirical risk. The paper next focuses on the formalization of an efficient methodology based on implicit differentiation that enables the introduction of the optimization of PAC-Bayesian risk certificates inside the loss/objective function used to fit the network/model. The last contribution is a method to optimize bounds on non-differentiable objectives such as the 0-1 loss. These theoretical contributions are complemented with an empirical evaluation on the MNIST and CIFAR-10 datasets. In fact, this paper presents the first non-vacuous generalization bounds on CIFAR-10 for neural networks. Code to reproduce all experiments is available at github.com/Diegogpcm/pacbayesgradients.

Sahan Bulathwela, María Pérez-Ortiz, Catherine Holloway et al.

Artificial Intelligence (AI) in Education has been said to have the potential for building more personalised curricula, as well as democratising education worldwide and creating a Renaissance of new ways of teaching and learning. Millions of students are already starting to benefit from the use of these technologies, but millions more around the world are not. If this trend continues, the first delivery of AI in Education could be greater educational inequality, along with a global misallocation of educational resources motivated by the current technological determinism narrative. In this paper, we focus on speculating and posing questions around the future of AI in Education, with the aim of starting the pressing conversation that would set the right foundations for the new generation of education that is permeated by technology. This paper starts by synthesising how AI might change how we learn and teach, focusing specifically on the case of personalised learning companions, and then move to discuss some socio-technical features that will be crucial for avoiding the perils of these AI systems worldwide (and perhaps ensuring their success). This paper also discusses the potential of using AI together with free, participatory and democratic resources, such as Wikipedia, Open Educational Resources and open-source tools. We also emphasise the need for collectively designing human-centered, transparent, interactive and collaborative AI-based algorithms that empower and give complete agency to stakeholders, as well as support new emerging pedagogies. Finally, we ask what would it take for this educational revolution to provide egalitarian and empowering access to education, beyond any political, cultural, language, geographical and learning ability barriers.

Sahan Bulathwela, Maria Perez-Ortiz, Erik Novak et al.

Educational recommenders have received much less attention in comparison to e-commerce and entertainment-related recommenders, even though efficient intelligent tutors have great potential to improve learning gains. One of the main challenges in advancing this research direction is the scarcity of large, publicly available datasets. In this work, we release a large, novel dataset of learners engaging with educational videos in-the-wild. The dataset, named Personalised Educational Engagement with Knowledge Topics PEEK, is the first publicly available dataset of this nature. The video lectures have been associated with Wikipedia concepts related to the material of the lecture, thus providing a humanly intuitive taxonomy. We believe that granular learner engagement signals in unison with rich content representations will pave the way to building powerful personalization algorithms that will revolutionise educational and informational recommendation systems. Towards this goal, we 1) construct a novel dataset from a popular video lecture repository, 2) identify a set of benchmark algorithms to model engagement, and 3) run extensive experimentation on the PEEK dataset to demonstrate its value. Our experiments with the dataset show promise in building powerful informational recommender systems. The dataset and the support code is available publicly.

Théophile Cantelobre, Benjamin Guedj, María Pérez-Ortiz et al.

Many practical machine learning tasks can be framed as Structured prediction problems, where several output variables are predicted and considered interdependent. Recent theoretical advances in structured prediction have focused on obtaining fast rates convergence guarantees, especially in the Implicit Loss Embedding (ILE) framework. PAC-Bayes has gained interest recently for its capacity of producing tight risk bounds for predictor distributions. This work proposes a novel PAC-Bayes perspective on the ILE Structured prediction framework. We present two generalization bounds, on the risk and excess risk, which yield insights into the behavior of ILE predictors. Two learning algorithms are derived from these bounds. The algorithms are implemented and their behavior analyzed, with source code available at \url{https://github.com/theophilec/PAC-Bayes-ILE-Structured-Prediction}.

Sahan Bulathwela, Maria Perez-Ortiz, Emine Yilmaz et al.

With the emergence of e-learning and personalised education, the production and distribution of digital educational resources have boomed. Video lectures have now become one of the primary modalities to impart knowledge to masses in the current digital age. The rapid creation of video lecture content challenges the currently established human-centred moderation and quality assurance pipeline, demanding for more efficient, scalable and automatic solutions for managing learning resources. Although a few datasets related to engagement with educational videos exist, there is still an important need for data and research aimed at understanding learner engagement with scientific video lectures. This paper introduces VLEngagement, a novel dataset that consists of content-based and video-specific features extracted from publicly available scientific video lectures and several metrics related to user engagement. We introduce several novel tasks related to predicting and understanding context-agnostic engagement in video lectures, providing preliminary baselines. This is the largest and most diverse publicly available dataset to our knowledge that deals with such tasks. The extraction of Wikipedia topic-based features also allows associating more sophisticated Wikipedia based features to the dataset to improve the performance in these tasks. The dataset, helper tools and example code snippets are available publicly at https://github.com/sahanbull/context-agnostic-engagement

Yuxiang Qiu, Karim Djemili, Denis Elezi et al.

With the advancement and utility of Artificial Intelligence (AI), personalising education to a global population could be a cornerstone of new educational systems in the future. This work presents the PEEKC dataset and the TrueLearn Python library, which contains a dataset and a series of online learner state models that are essential to facilitate research on learner engagement modelling.TrueLearn family of models was designed following the "open learner" concept, using humanly-intuitive user representations. This family of scalable, online models also help end-users visualise the learner models, which may in the future facilitate user interaction with their models/recommenders. The extensive documentation and coding examples make the library highly accessible to both machine learning developers and educational data mining and learning analytics practitioners. The experiments show the utility of both the dataset and the library with predictive performance significantly exceeding comparative baseline models. The dataset contains a large amount of AI-related educational videos, which are of interest for building and validating AI-specific educational recommenders.

Filip Ilievski, Barbara Hammer, Frank van Harmelen et al.

Recent advances in AI -- including generative approaches -- have resulted in technology that can support humans in scientific discovery and forming decisions, but may also disrupt democracies and target individuals. The responsible use of AI and its participation in human-AI teams increasingly shows the need for AI alignment, that is, to make AI systems act according to our preferences. A crucial yet often overlooked aspect of these interactions is the different ways in which humans and machines generalise. In cognitive science, human generalisation commonly involves abstraction and concept learning. In contrast, AI generalisation encompasses out-of-domain generalisation in machine learning, rule-based reasoning in symbolic AI, and abstraction in neurosymbolic AI. In this perspective paper, we combine insights from AI and cognitive science to identify key commonalities and differences across three dimensions: notions of, methods for, and evaluation of generalisation. We map the different conceptualisations of generalisation in AI and cognitive science along these three dimensions and consider their role for alignment in human-AI teaming. This results in interdisciplinary challenges across AI and cognitive science that must be tackled to provide a foundation for effective and cognitively supported alignment in human-AI teaming scenarios.

Theodore Wolf, Nantas Nardelli, John Shawe-Taylor et al.

Governments around the world aspire to ground decision-making on evidence. Many of the foundations of policy making - e.g. sensing patterns that relate to societal needs, developing evidence-based programs, forecasting potential outcomes of policy changes, and monitoring effectiveness of policy programs - have the potential to benefit from the use of large-scale datasets or simulations together with intelligent algorithms. These could, if designed and deployed in a way that is well grounded on scientific evidence, enable a more comprehensive, faster, and rigorous approach to policy making. Integrated Assessment Models (IAM) is a broad umbrella covering scientific models that attempt to link main features of society and economy with the biosphere into one modelling framework. At present, these systems are probed by policy makers and advisory groups in a hypothesis-driven manner. In this paper, we empirically demonstrate that modern Reinforcement Learning can be used to probe IAMs and explore the space of solutions in a more principled manner. While the implication of our results are modest since the environment is simplistic, we believe that this is a stepping stone towards more ambitious use cases, which could allow for effective exploration of policies and understanding of their consequences and limitations.

Simon Schmitt, John Shawe-Taylor, Hado van Hasselt

Decision makers may suffer from uncertainty induced by limited data. This may be mitigated by accounting for epistemic uncertainty, which is however challenging to estimate efficiently for large neural networks. To this extent we investigate Delta Variances, a family of algorithms for epistemic uncertainty quantification, that is computationally efficient and convenient to implement. It can be applied to neural networks and more general functions composed of neural networks. As an example we consider a weather simulator with a neural-network-based step function inside -- here Delta Variances empirically obtain competitive results at the cost of a single gradient computation. The approach is convenient as it requires no changes to the neural network architecture or training procedure. We discuss multiple ways to derive Delta Variances theoretically noting that special cases recover popular techniques and present a unified perspective on multiple related methods. Finally we observe that this general perspective gives rise to a natural extension and empirically show its benefit.

Sahan Bulathwela, Daniel Van Niekerk, Jarrod Shipton et al.

Personalised education is one of the domains that can greatly benefit from the most recent advances in Artificial Intelligence (AI) and Large Language Models (LLM). However, it is also one of the most challenging applications due to the cognitive complexity of teaching effectively while personalising the learning experience to suit independent learners. We hypothesise that one promising approach to excelling in such demanding use cases is using a \emph{society of minds}. In this chapter, we present TrueReason, an exemplar personalised learning system that integrates a multitude of specialised AI models that can mimic micro skills that are composed together by a LLM to operationalise planning and reasoning. The architecture of the initial prototype is presented while describing two micro skills that have been incorporated in the prototype. The proposed system demonstrates the first step in building sophisticated AI systems that can take up very complex cognitive tasks that are demanded by domains such as education.

Yuta Nagano, Andrew Pyo, Martina Milighetti et al.

Computational prediction of the interaction of T cell receptors (TCRs) and their ligands is a grand challenge in immunology. Despite advances in high-throughput assays, specificity-labelled TCR data remains sparse. In other domains, the pre-training of language models on unlabelled data has been successfully used to address data bottlenecks. However, it is unclear how to best pre-train protein language models for TCR specificity prediction. Here we introduce a TCR language model called SCEPTR (Simple Contrastive Embedding of the Primary sequence of T cell Receptors), capable of data-efficient transfer learning. Through our model, we introduce a novel pre-training strategy combining autocontrastive learning and masked-language modelling, which enables SCEPTR to achieve its state-of-the-art performance. In contrast, existing protein language models and a variant of SCEPTR pre-trained without autocontrastive learning are outperformed by sequence alignment-based methods. We anticipate that contrastive learning will be a useful paradigm to decode the rules of TCR specificity.

Reuben Adams, John Shawe-Taylor, Benjamin Guedj

Current PAC-Bayes generalisation bounds are restricted to scalar metrics of performance, such as the loss or error rate. However, one ideally wants more information-rich certificates that control the entire distribution of possible outcomes, such as the distribution of the test loss in regression, or the probabilities of different mis-classifications. We provide the first PAC-Bayes bound capable of providing such rich information by bounding the Kullback-Leibler divergence between the empirical and true probabilities of a set of $M$ error types, which can either be discretized loss values for regression, or the elements of the confusion matrix (or a partition thereof) for classification. We transform our bound into a differentiable training objective. Our bound is especially useful in cases where the severity of different mis-classifications may change over time; existing PAC-Bayes bounds can only bound a particular pre-decided weighting of the error types. In contrast our bound implicitly controls all uncountably many weightings simultaneously.

Simon Schmitt, John Shawe-Taylor, Hado van Hasselt

To accumulate knowledge and improve its policy of behaviour, a reinforcement learning agent can learn `off-policy' about policies that differ from the policy used to generate its experience. This is important to learn counterfactuals, or because the experience was generated out of its own control. However, off-policy learning is non-trivial, and standard reinforcement-learning algorithms can be unstable and divergent. In this paper we discuss a novel family of off-policy prediction algorithms which are convergent by construction. The idea is to first learn on-policy about the data-generating behaviour, and then bootstrap an off-policy value estimate on this on-policy estimate, thereby constructing a value estimate that is partially off-policy. This process can be repeated to build a chain of value functions, each time bootstrapping a new estimate on the previous estimate in the chain. Each step in the chain is stable and hence the complete algorithm is guaranteed to be stable. Under mild conditions this comes arbitrarily close to the off-policy TD solution when we increase the length of the chain. Hence it can compute the solution even in cases where off-policy TD diverges. We prove that the proposed scheme is convergent and corresponds to an iterative decomposition of the inverse key matrix. Furthermore it can be interpreted as estimating a novel objective -- that we call a `k-step expedition' -- of following the target policy for finitely many steps before continuing indefinitely with the behaviour policy. Empirically we evaluate the idea on challenging MDPs such as Baird's counter example and observe favourable results.

Maria Perez-Ortiz, Sahan Bulathwela, Claire Dormann et al.

Prior research has shown how 'content preview tools' improve speed and accuracy of user relevance judgements across different information retrieval tasks. This paper describes a novel user interface tool, the Content Flow Bar, designed to allow users to quickly identify relevant fragments within informational videos to facilitate browsing, through a cognitively augmented form of navigation. It achieves this by providing semantic "snippets" that enable the user to rapidly scan through video content. The tool provides visually-appealing pop-ups that appear in a time series bar at the bottom of each video, allowing to see in advance and at a glance how topics evolve in the content. We conducted a user study to evaluate how the tool changes the users search experience in video retrieval, as well as how it supports exploration and information seeking. The user questionnaire revealed that participants found the Content Flow Bar helpful and enjoyable for finding relevant information in videos. The interaction logs of the user study, where participants interacted with the tool for completing two informational tasks, showed that it holds promise for enhancing discoverability of content both across and within videos. This discovered potential could leverage a new generation of navigation tools in search and information retrieval.

Sahan Bulathwela, María Pérez-Ortiz, Emine Yilmaz et al.

In informational recommenders, many challenges arise from the need to handle the semantic and hierarchical structure between knowledge areas. This work aims to advance towards building a state-aware educational recommendation system that incorporates semantic relatedness between knowledge topics, propagating latent information across semantically related topics. We introduce a novel learner model that exploits this semantic relatedness between knowledge components in learning resources using the Wikipedia link graph, with the aim to better predict learner engagement and latent knowledge in a lifelong learning scenario. In this sense, Semantic TrueLearn builds a humanly intuitive knowledge representation while leveraging Bayesian machine learning to improve the predictive performance of the educational engagement. Our experiments with a large dataset demonstrate that this new semantic version of TrueLearn algorithm achieves statistically significant improvements in terms of predictive performance with a simple extension that adds semantic awareness to the model.

Maria Perez-Ortiz, Erik Novak, Sahan Bulathwela et al.

Artifical Intelligence (AI) in Education has great potential for building more personalised curricula, as well as democratising education worldwide and creating a Renaissance of new ways of teaching and learning. We believe this is a crucial moment for setting the foundations of AI in education in the beginning of this Fourth Industrial Revolution. This report aims to synthesize how AI might change (and is already changing) how we learn, as well as what technological features are crucial for these AI systems in education, with the end goal of starting this pressing dialogue of how the future of AI in education should unfold, engaging policy makers, engineers, researchers and obviously, teachers and learners. This report also presents the advances within the X5GON project, a European H2020 project aimed at building and deploying a cross-modal, cross-lingual, cross-cultural, cross-domain and cross-site personalised learning platform for Open Educational Resources (OER).

Maria Perez-Ortiz, Omar Rivasplata, Emilio Parrado-Hernandez et al.

A learning method is self-certified if it uses all available data to simultaneously learn a predictor and certify its quality with a tight statistical certificate that is valid on unseen data. Recent work has shown that neural network models trained by optimising PAC-Bayes bounds lead not only to accurate predictors, but also to tight risk certificates, bearing promise towards achieving self-certified learning. In this context, learning and certification strategies based on PAC-Bayes bounds are especially attractive due to their ability to leverage all data to learn a posterior and simultaneously certify its risk with a tight numerical certificate. In this paper, we assess the progress towards self-certification in probabilistic neural networks learnt by PAC-Bayes inspired objectives. We empirically compare (on 4 classification datasets) classical test set bounds for deterministic predictors and a PAC-Bayes bound for randomised self-certified predictors. We first show that both of these generalisation bounds are not too far from out-of-sample test set errors. We then show that in data starvation regimes, holding out data for the test set bounds adversely affects generalisation performance, while self-certified strategies based on PAC-Bayes bounds do not suffer from this drawback, proving that they might be a suitable choice for the small data regime. We also find that probabilistic neural networks learnt by PAC-Bayes inspired objectives lead to certificates that can be surprisingly competitive with commonly used test set bounds.

Maria Perez-Ortiz, Omar Rivasplata, Benjamin Guedj et al.

Recent works have investigated deep learning models trained by optimising PAC-Bayes bounds, with priors that are learnt on subsets of the data. This combination has been shown to lead not only to accurate classifiers, but also to remarkably tight risk certificates, bearing promise towards self-certified learning (i.e. use all the data to learn a predictor and certify its quality). In this work, we empirically investigate the role of the prior. We experiment on 6 datasets with different strategies and amounts of data to learn data-dependent PAC-Bayes priors, and we compare them in terms of their effect on test performance of the learnt predictors and tightness of their risk certificate. We ask what is the optimal amount of data which should be allocated for building the prior and show that the optimum may be dataset dependent. We demonstrate that using a small percentage of the prior-building data for validation of the prior leads to promising results. We include a comparison of underparameterised and overparameterised models, along with an empirical study of different training objectives and regularisation strategies to learn the prior distribution.

Maxime Haddouche, Benjamin Guedj, John Shawe-Taylor

Principal Component Analysis (PCA) is a popular method for dimension reduction and has attracted an unfailing interest for decades. More recently, kernel PCA (KPCA) has emerged as an extension of PCA but, despite its use in practice, a sound theoretical understanding of KPCA is missing. We contribute several lower and upper bounds on the efficiency of KPCA, involving the empirical eigenvalues of the kernel Gram matrix and new quantities involving a notion of variance. These bounds show how much information is captured by KPCA on average and contribute a better theoretical understanding of its efficiency. We demonstrate that fast convergence rates are achievable for a widely used class of kernels and we highlight the importance of some desirable properties of datasets to ensure KPCA efficiency.

María Pérez-Ortiz, Omar Rivasplata, John Shawe-Taylor et al.

This paper presents an empirical study regarding training probabilistic neural networks using training objectives derived from PAC-Bayes bounds. In the context of probabilistic neural networks, the output of training is a probability distribution over network weights. We present two training objectives, used here for the first time in connection with training neural networks. These two training objectives are derived from tight PAC-Bayes bounds. We also re-implement a previously used training objective based on a classical PAC-Bayes bound, to compare the properties of the predictors learned using the different training objectives. We compute risk certificates for the learnt predictors, based on part of the data used to learn the predictors. We further experiment with different types of priors on the weights (both data-free and data-dependent priors) and neural network architectures. Our experiments on MNIST and CIFAR-10 show that our training methods produce competitive test set errors and non-vacuous risk bounds with much tighter values than previous results in the literature, showing promise not only to guide the learning algorithm through bounding the risk but also for model selection. These observations suggest that the methods studied here might be good candidates for self-certified learning, in the sense of using the whole data set for learning a predictor and certifying its risk on any unseen data (from the same distribution as the training data) potentially without the need for holding out test data.

Omar Rivasplata, Ilja Kuzborskij, Csaba Szepesvari et al.

We focus on a stochastic learning model where the learner observes a finite set of training examples and the output of the learning process is a data-dependent distribution over a space of hypotheses. The learned data-dependent distribution is then used to make randomized predictions, and the high-level theme addressed here is guaranteeing the quality of predictions on examples that were not seen during training, i.e. generalization. In this setting the unknown quantity of interest is the expected risk of the data-dependent randomized predictor, for which upper bounds can be derived via a PAC-Bayes analysis, leading to PAC-Bayes bounds. Specifically, we present a basic PAC-Bayes inequality for stochastic kernels, from which one may derive extensions of various known PAC-Bayes bounds as well as novel bounds. We clarify the role of the requirements of fixed 'data-free' priors, bounded losses, and i.i.d. data. We highlight that those requirements were used to upper-bound an exponential moment term, while the basic PAC-Bayes theorem remains valid without those restrictions. We present three bounds that illustrate the use of data-dependent priors, including one for the unbounded square loss.

Maxime Haddouche, Benjamin Guedj, Omar Rivasplata et al.

We present new PAC-Bayesian generalisation bounds for learning problems with unbounded loss functions. This extends the relevance and applicability of the PAC-Bayes learning framework, where most of the existing literature focuses on supervised learning problems with a bounded loss function (typically assumed to take values in the interval [0;1]). In order to relax this assumption, we propose a new notion called HYPE (standing for \emph{HYPothesis-dependent rangE}), which effectively allows the range of the loss to depend on each predictor. Based on this new notion we derive a novel PAC-Bayesian generalisation bound for unbounded loss functions, and we instantiate it on a linear regression problem. To make our theory usable by the largest audience possible, we include discussions on actual computation, practicality and limitations of our assumptions.

Sahan Bulathwela, María Pérez-Ortiz, Aldo Lipani et al.

The explosion of Open Educational Resources (OERs) in the recent years creates the demand for scalable, automatic approaches to process and evaluate OERs, with the end goal of identifying and recommending the most suitable educational materials for learners. We focus on building models to find the characteristics and features involved in context-agnostic engagement (i.e. population-based), a seldom researched topic compared to other contextualised and personalised approaches that focus more on individual learner engagement. Learner engagement, is arguably a more reliable measure than popularity/number of views, is more abundant than user ratings and has also been shown to be a crucial component in achieving learning outcomes. In this work, we explore the idea of building a predictive model for population-based engagement in education. We introduce a novel, large dataset of video lectures for predicting context-agnostic engagement and propose both cross-modal and modality-specific feature sets to achieve this task. We further test different strategies for quantifying learner engagement signals. We demonstrate the use of our approach in the case of data scarcity. Additionally, we perform a sensitivity analysis of the best performing model, which shows promising performance and can be easily integrated into an educational recommender system for OERs.

Yuxin Sun, Benny Chain, Samuel Kaski et al.

In many high dimensional classification or regression problems set in a biological context, the complete identification of the set of informative features is often as important as predictive accuracy, since this can provide mechanistic insight and conceptual understanding. Lasso and related algorithms have been widely used since their sparse solutions naturally identify a set of informative features. However, Lasso performs erratically when features are correlated. This limits the use of such algorithms in biological problems, where features such as genes often work together in pathways, leading to sets of highly correlated features. In this paper, we examine the performance of a Lasso derivative, the exclusive group Lasso, in this setting. We propose fast algorithms to solve the exclusive group Lasso, and introduce a solution to the case when the underlying group structure is unknown. The solution combines stability selection with random group allocation and introduction of artificial features. Experiments with both synthetic and real-world data highlight the advantages of this proposed methodology over Lasso in comprehensive selection of informative features.

Sahan Bulathwela, Maria Perez-Ortiz, Emine Yilmaz et al.

One of the most ambitious use cases of computer-assisted learning is to build a recommendation system for lifelong learning. Most recommender algorithms exploit similarities between content and users, overseeing the necessity to leverage sensible learning trajectories for the learner. Lifelong learning thus presents unique challenges, requiring scalable and transparent models that can account for learner knowledge and content novelty simultaneously, while also retaining accurate learners representations for long periods of time. We attempt to build a novel educational recommender, that relies on an integrative approach combining multiple drivers of learners engagement. Our first step towards this goal is TrueLearn, which models content novelty and background knowledge of learners and achieves promising performance while retaining a human interpretable learner model.

Sahan Bulathwela, Maria Perez-Ortiz, Emine Yilmaz et al.

The recent advances in computer-assisted learning systems and the availability of open educational resources today promise a pathway to providing cost-efficient, high-quality education to large masses of learners. One of the most ambitious use cases of computer-assisted learning is to build a lifelong learning recommendation system. Unlike short-term courses, lifelong learning presents unique challenges, requiring sophisticated recommendation models that account for a wide range of factors such as background knowledge of learners or novelty of the material while effectively maintaining knowledge states of masses of learners for significantly longer periods of time (ideally, a lifetime). This work presents the foundations towards building a dynamic, scalable and transparent recommendation system for education, modelling learner's knowledge from implicit data in the form of engagement with open educational resources. We i) use a text ontology based on Wikipedia to automatically extract knowledge components of educational resources and, ii) propose a set of online Bayesian strategies inspired by the well-known areas of item response theory and knowledge tracing. Our proposal, TrueLearn, focuses on recommendations for which the learner has enough background knowledge (so they are able to understand and learn from the material), and the material has enough novelty that would help the learner improve their knowledge about the subject and keep them engaged. We further construct a large open educational video lectures dataset and test the performance of the proposed algorithms, which show clear promise towards building an effective educational recommendation system.

Gaurav Singh, Zahra Sabet, John Shawe-Taylor et al.

Biomedical text tagging systems are plagued by the dearth of labeled training data. There have been recent attempts at using pre-trained encoders to deal with this issue. Pre-trained encoder provides representation of the input text which is then fed to task-specific layers for classification. The entire network is fine-tuned on the labeled data from the target task. Unfortunately, a low-resource biomedical task often has too few labeled instances for satisfactory fine-tuning. Also, if the label space is large, it contains few or no labeled instances for majority of the labels. Most biomedical tagging systems treat labels as indexes, ignoring the fact that these labels are often concepts expressed in natural language e.g. `Appearance of lesion on brain imaging'. To address these issues, we propose constructing extra labeled instances using label-text (i.e. label's name) as input for the corresponding label-index (i.e. label's index). In fact, we propose a number of strategies for manufacturing multiple artificial labeled instances from a single label. The network is then fine-tuned on a combination of real and these newly constructed artificial labeled instances. We evaluate the proposed approach on an important low-resource biomedical task called \textit{PICO annotation}, which requires tagging raw text describing clinical trials with labels corresponding to different aspects of the trial i.e. PICO (Population, Intervention/Control, Outcome) characteristics of the trial. Our empirical results show that the proposed method achieves a new state-of-the-art performance for PICO annotation with very significant improvements over competitive baselines.

Biobele J. Brown, Alexander A. Przybylski, Petru Manescu et al.

Plasmodium falciparum malaria still poses one of the greatest threats to human life with over 200 million cases globally leading to half-million deaths annually. Of these, 90% of cases and of the mortality occurs in sub-Saharan Africa, mostly among children. Although malaria prediction systems are central to the 2016-2030 malaria Global Technical Strategy, currently these are inadequate at capturing and estimating the burden of disease in highly endemic countries. We developed and validated a computational system that exploits the predictive power of current Machine Learning approaches on 22-years of prospective data from the high-transmission holoendemic malaria urban-densely-populated sub-Saharan West-Africa metropolis of Ibadan. Our dataset of >9x104 screened study participants attending our clinical and community services from 1996 to 2017 contains monthly prevalence, temporal, environmental and host features. Our Locality-specific Elastic-Net based Malaria Prediction System (LEMPS) achieves good generalization performance, both in magnitude and direction of the prediction, when tasked to predict monthly prevalence on previously unseen validation data (MAE<=6x10-2, MSE<=7x10-3) within a range of (+0.1 to -0.05) error-tolerance which is relevant and usable for aiding decision-support in a holoendemic setting. LEMPS is well-suited for malaria prediction, where there are multiple features which are correlated with one another, and trading-off between regularization-strength L1-norm and L2-norm allows the system to retain stability. Data-driven systems are critical for regionally-adaptable surveillance, management of control strategies and resource allocation across stretched healthcare systems.

Petru Manescu, Lydia Neary- Zajiczek, Michael J. Shaw et al.

Fast accurate diagnosis of malaria is still a global health challenge for which automated digital-pathology approaches could provide scalable solutions amenable to be deployed in low-to-middle income countries. Here we address the problem of Extended Depth-of-Field (EDoF) in thick blood film microscopy for rapid automated malaria diagnosis. High magnification oil-objectives (100x) with large numerical aperture are usually preferred to resolve the fine structural details that help separate true parasites from distractors. However, such objectives have a very limited depth-of-field requiring the acquisition of a series of images at different focal planes per field of view (FOV). Current EDoF techniques based on multi-scale decompositions are time consuming and therefore not suited for high-throughput analysis of specimens. To overcome this challenge, we developed a new deep learning method based on Convolutional Neural Networks (EDoF-CNN) that is able to rapidly perform the extended depth-of-field while also enhancing the spatial resolution of the resulting fused image. We evaluated our approach using simulated low-resolution z-stacks from Giemsa-stained thick blood smears from patients presenting with Plasmodium falciparum malaria. The EDoF-CNN allows speed-up of our digital-pathology acquisition platform and significantly improves the quality of the EDoF compared to the traditional multi-scaled approaches when applied to lower resolution stacks corresponding to acquisitions with fewer focal planes, large camera pixel binning or lower magnification objectives (larger FOV). We use the parasite detection accuracy of a deep learning model on the EDoFs as a concrete, task-specific measure of performance of this approach.

Jie M. Zhang, Mark Harman, Benjamin Guedj et al.

We introduce an exploratory study on Mutation Validation (MV), a model validation method using mutated training labels for supervised learning. MV mutates training data labels, retrains the model against the mutated data, then uses the metamorphic relation that captures the consequent training performance changes to assess model fit. It does not use a validation set or test set. The intuition underpinning MV is that overfitting models tend to fit noise in the training data. We explore 8 different learning algorithms, 18 datasets, and 5 types of hyperparameter tuning tasks. Our results demonstrate that MV is accurate in model selection: the model recommendation hit rate is 92\% for MV and less than 60\% for out-of-sample-validation. MV also provides more stable hyperparameter tuning results than out-of-sample-validation across different runs.

Luke R Harries, Suyi Zhang, Geoffroy Dubourg-Felonneau et al.

DNA sequencing to identify genetic variants is becoming increasingly valuable in clinical settings. Assessment of variants in such sequencing data is commonly implemented through Bayesian heuristic algorithms. Machine learning has shown great promise in improving on these variant calls, but the input for these is still a standardized "pile-up" image, which is not always best suited. In this paper, we present a novel method for generating images from DNA sequencing data, which interlaces the human reference genome with personalized sequencing output, to maximize usage of sequencing reads and improve machine learning algorithm performance. We demonstrate the success of this in improving standard germline variant calling. We also furthered this approach to include somatic variant calling across tumor/normal data with Siamese networks. These approaches can be used in machine learning applications on sequencing data with the hope of improving clinical outcomes, and are freely available for noncommercial use at www.ccg.ai.

Gaurav Singh, James Thomas, Iain J. Marshall et al.

We propose a model for tagging unstructured texts with an arbitrary number of terms drawn from a tree-structured vocabulary (i.e., an ontology). We treat this as a special case of sequence-to-sequence learning in which the decoder begins at the root node of an ontological tree and recursively elects to expand child nodes as a function of the input text, the current node, and the latent decoder state. In our experiments the proposed method outperforms state-of-the-art approaches on the important task of automatically assigning MeSH terms to biomedical abstracts.

Gaurav Singh, John Shawe-Taylor

Deep neural networks have gained tremendous popularity in last few years. They have been applied for the task of classification in almost every domain. Despite the success, deep networks can be incredibly slow to train for even moderate sized models on sufficiently large datasets. Additionally, these networks require large amounts of data to be able to generalize. The importance of speeding up convergence, and generalization in deep networks can not be overstated. In this work, we develop an optimization algorithm based on generalized-optimal updates derived from minibatches that lead to faster convergence. Towards the end, we demonstrate on two benchmark datasets that the proposed method achieves two orders of magnitude speed up over traditional back-propagation, and is more robust to noise/over-fitting.

Seth Nabarro, Tristan Fletcher, John Shawe-Taylor

Accurately predicting when and where ambulance call-outs occur can reduce response times and ensure the patient receives urgent care sooner. Here we present a novel method for ambulance demand prediction using Gaussian process regression (GPR) in time and geographic space. The method exhibits superior accuracy to MEDIC, a method which has been used in industry. The use of GPR has additional benefits such as the quantification of uncertainty with each prediction, the choice of kernel functions to encode prior knowledge and the ability to capture spatial correlation. Measures to increase the utility of GPR in the current context, with large training sets and a Poisson-distributed output, are outlined.

Omar Rivasplata, Emilio Parrado-Hernandez, John Shawe-Taylor et al.

PAC-Bayes bounds have been proposed to get risk estimates based on a training sample. In this paper the PAC-Bayes approach is combined with stability of the hypothesis learned by a Hilbert space valued algorithm. The PAC-Bayes setting is used with a Gaussian prior centered at the expected output. Thus a novelty of our paper is using priors defined in terms of the data-generating distribution. Our main result estimates the risk of the randomized algorithm in terms of the hypothesis stability coefficients. We also provide a new bound for the SVM classifier, which is compared to other known bounds experimentally. Ours appears to be the first stability-based bound that evaluates to non-trivial values.

Tobias Baumann, Thore Graepel, John Shawe-Taylor

In the future, artificial learning agents are likely to become increasingly widespread in our society. They will interact with both other learning agents and humans in a variety of complex settings including social dilemmas. We consider the problem of how an external agent can promote cooperation between artificial learners by distributing additional rewards and punishments based on observing the learners' actions. We propose a rule for automatically learning how to create right incentives by considering the players' anticipated parameter updates. Using this learning rule leads to cooperation with high social welfare in matrix games in which the agents would otherwise learn to defect with high probability. We show that the resulting cooperative outcome is stable in certain games even if the planning agent is turned off after a given number of episodes, while other games require ongoing intervention to maintain mutual cooperation. However, even in the latter case, the amount of necessary additional incentives decreases over time.

Michele Donini, Luca Oneto, Shai Ben-David et al.

We address the problem of algorithmic fairness: ensuring that sensitive variables do not unfairly influence the outcome of a classifier. We present an approach based on empirical risk minimization, which incorporates a fairness constraint into the learning problem. It encourages the conditional risk of the learned classifier to be approximately constant with respect to the sensitive variable. We derive both risk and fairness bounds that support the statistical consistency of our approach. We specify our approach to kernel methods and observe that the fairness requirement implies an orthogonality constraint which can be easily added to these methods. We further observe that for linear models the constraint translates into a simple data preprocessing step. Experiments indicate that the method is empirically effective and performs favorably against state-of-the-art approaches.

Gaurav Singh, James Thomas, John Shawe-Taylor

Systematic reviews are essential to summarizing the results of different clinical and social science studies. The first step in a systematic review task is to identify all the studies relevant to the review. The task of identifying relevant studies for a given systematic review is usually performed manually, and as a result, involves substantial amounts of expensive human resource. Lately, there have been some attempts to reduce this manual effort using active learning. In this work, we build upon some such existing techniques, and validate by experimenting on a larger and comprehensive dataset than has been attempted until now. Our experiments provide insights on the use of different feature extraction models for different disciplines. More importantly, we identify that a naive active learning based screening process is biased in favour of selecting similar documents. We aimed to improve the performance of the screening process using a novel active learning algorithm with success. Additionally, we propose a mechanism to choose the best feature extraction method for a given review.

Viivi Uurtio, João M. Monteiro, Jaz Kandola et al.

Canonical correlation analysis is a family of multivariate statistical methods for the analysis of paired sets of variables. Since its proposition, canonical correlation analysis has for instance been extended to extract relations between two sets of variables when the sample size is insufficient in relation to the data dimensionality, when the relations have been considered to be non-linear, and when the dimensionality is too large for human interpretation. This tutorial explains the theory of canonical correlation analysis including its regularised, kernel, and sparse variants. Additionally, the deep and Bayesian CCA extensions are briefly reviewed. Together with the numerical examples, this overview provides a coherent compendium on the applicability of the variants of canonical correlation analysis. By bringing together techniques for solving the optimisation problems, evaluating the statistical significance and generalisability of the canonical correlation model, and interpreting the relations, we hope that this article can serve as a hands-on tool for applying canonical correlation methods in data analysis.

Kira Kempinska, Toby Davies, John Shawe-Taylor

Increasing availability of vehicle GPS data has created potentially transformative opportunities for traffic management, route planning and other location-based services. Critical to the utility of the data is their accuracy. Map-matching is the process of improving the accuracy by aligning GPS data with the road network. In this paper, we propose a purely probabilistic approach to map-matching based on a sequential Monte Carlo algorithm known as particle filters. The approach performs map-matching by producing a range of candidate solutions, each with an associated probability score. We outline implementation details and thoroughly validate the technique on GPS data of varied quality.

Kira Kempinska, John Shawe-Taylor

The ability to track a moving vehicle is of crucial importance in numerous applications. The task has often been approached by the importance sampling technique of particle filters due to its ability to model non-linear and non-Gaussian dynamics, of which a vehicle travelling on a road network is a good example. Particle filters perform poorly when observations are highly informative. In this paper, we address this problem by proposing particle filters that sample around the most recent observation. The proposal leads to an order of magnitude improvement in accuracy and efficiency over conventional particle filters, especially when observations are infrequent but low-noise.

Gaurav Singh, Fabrizio Silvestri, John Shawe-Taylor

In a traditional setting, classifiers are trained to approximate a target function $f:X \rightarrow Y$ where at least a sample for each $y \in Y$ is presented to the training algorithm. In a zero-shot setting we have a subset of the labels $\hat{Y} \subset Y$ for which we do not observe any corresponding training instance. Still, the function $f$ that we train must be able to correctly assign labels also on $\hat{Y}$. In practice, zero-shot problems are very important especially when the label set is large and the cost of editorially label samples for all possible values in the label set might be prohibitively high. Most recent approaches to zero-shot learning are based on finding and exploiting relationships between labels using semantic embeddings. We show in this paper that semantic embeddings, despite being very good at capturing relationships between labels, are not very good at capturing the relationships among labels in a data-dependent manner. For this reason, we propose a novel two-step process for learning a zero-shot classifier. In the first step, we learn what we call a \emph{property embedding space} capturing the "\emph{learnable}" features of the label set. Then, we exploit the learned properties in order to reduce the generalization error for a linear nearest neighbor-based classifier.

Dorota Glowacka, Yee Whye Teh, John Shawe-Taylor

A content-based image retrieval system based on multinomial relevance feedback is proposed. The system relies on an interactive search paradigm where at each round a user is presented with k images and selects the one closest to their ideal target. Two approaches, one based on the Dirichlet distribution and one based the Beta distribution, are used to model the problem motivating an algorithm that trades exploration and exploitation in presenting the images in each round. Experimental results show that the new approach compares favourably with previous work.

Makoto Yamada, Koh Takeuchi, Tomoharu Iwata et al.

We introduce the localized Lasso, which is suited for learning models that are both interpretable and have a high predictive power in problems with high dimensionality $d$ and small sample size $n$. More specifically, we consider a function defined by local sparse models, one at each data point. We introduce sample-wise network regularization to borrow strength across the models, and sample-wise exclusive group sparsity (a.k.a., $\ell_{1,2}$ norm) to introduce diversity into the choice of feature sets in the local models. The local models are interpretable in terms of similarity of their sparsity patterns. The cost function is convex, and thus has a globally optimal solution. Moreover, we propose a simple yet efficient iterative least-squares based optimization procedure for the localized Lasso, which does not need a tuning parameter, and is guaranteed to converge to a globally optimal solution. The solution is empirically shown to outperform alternatives for both simulated and genomic personalized medicine data.

Diana Borsa, Thore Graepel, John Shawe-Taylor

We investigate a paradigm in multi-task reinforcement learning (MT-RL) in which an agent is placed in an environment and needs to learn to perform a series of tasks, within this space. Since the environment does not change, there is potentially a lot of common ground amongst tasks and learning to solve them individually seems extremely wasteful. In this paper, we explicitly model and learn this shared structure as it arises in the state-action value space. We will show how one can jointly learn optimal value-functions by modifying the popular Value-Iteration and Policy-Iteration procedures to accommodate this shared representation assumption and leverage the power of multi-task supervised learning. Finally, we demonstrate that the proposed model and training procedures, are able to infer good value functions, even under low samples regimes. In addition to data efficiency, we will show in our analysis, that learning abstractions of the state space jointly across tasks leads to more robust, transferable representations with the potential for better generalization. this shared representation assumption and leverage the power of multi-task supervised learning. Finally, we demonstrate that the proposed model and training procedures, are able to infer good value functions, even under low samples regimes. In addition to data efficiency, we will show in our analysis, that learning abstractions of the state space jointly across tasks leads to more robust, transferable representations with the potential for better generalization.