Predicting the behaviour of shoppers provides valuable information for retailers, such as the expected spend of a shopper or the total turnover of a supermarket. The ability to make predictions on an individual level is useful, as it allows supermarkets to accurately perform targeted marketing. However, given the expected number of shoppers and their diverse behaviours, making accurate predictions on an individual level is difficult. This problem does not only arise in shopper behaviour, but also in various business processes, such as predicting when an invoice will be paid. In this paper we present CAPiES, a framework that focuses on this trade-off in an online setting. By making predictions on a larger number of entities at a time, we improve the predictive accuracy but at the potential cost of usefulness since we can say less about the individual entities. CAPiES is developed in an online setting, where we continuously update the prediction model and make new predictions over time. We show the existence of the trade-off in an experimental evaluation in two real-world scenarios: a supermarket with over 160 000 shoppers and a paint factory with over 171 000 invoices.

Contrastive self-supervised learning (CSL) with a prototypical regularization has been introduced in learning meaningful representations for downstream tasks that require strong semantic information. However, to optimize CSL with a loss that performs the prototypical regularization aggressively, e.g., the ProtoNCE loss, might cause the "coagulation" of examples in the embedding space. That is, the intra-prototype diversity of samples collapses to trivial solutions for their prototype being well-separated from others. Motivated by previous works, we propose to mitigate this phenomenon by learning Prototypical representation through Alignment, Uniformity and Correlation (PAUC). Specifically, the ordinary ProtoNCE loss is revised with: (1) an alignment loss that pulls embeddings from positive prototypes together; (2) a uniformity loss that distributes the prototypical level features uniformly; (3) a correlation loss that increases the diversity and discriminability between prototypical level features. We conduct extensive experiments on various benchmarks where the results demonstrate the effectiveness of our method in improving the quality of prototypical contrastive representations. Particularly, in the classification down-stream tasks with linear probes, our proposed method outperforms the state-of-the-art instance-wise and prototypical contrastive learning methods on the ImageNet-100 dataset by 2.96% and the ImageNet-1K dataset by 2.46% under the same settings of batch size and epochs.

Equity is widely held to be fundamental to the ethics of healthcare. In the context of clinical decision-making, it rests on the comparative fidelity of the intelligence -- evidence-based or intuitive -- guiding the management of each individual patient. Though brought to recent attention by the individuating power of contemporary machine learning, such epistemic equity arises in the context of any decision guidance, whether traditional or innovative. Yet no general framework for its quantification, let alone assurance, currently exists. Here we formulate epistemic equity in terms of model fidelity evaluated over learnt multi-dimensional representations of identity crafted to maximise the captured diversity of the population, introducing a comprehensive framework for Representational Ethical Model Calibration. We demonstrate use of the framework on large-scale multimodal data from UK Biobank to derive diverse representations of the population, quantify model performance, and institute responsive remediation. We offer our approach as a principled solution to quantifying and assuring epistemic equity in healthcare, with applications across the research, clinical, and regulatory domains.

Generalized Additive Models (GAMs) have quickly become the leading choice for inherently-interpretable machine learning. However, unlike uninterpretable methods such as DNNs, they lack expressive power and easy scalability, and are hence not a feasible alternative for real-world tasks. We present a new class of GAMs that use tensor rank decompositions of polynomials to learn powerful, {\em inherently-interpretable} models. Our approach, titled Scalable Polynomial Additive Models (SPAM) is effortlessly scalable and models {\em all} higher-order feature interactions without a combinatorial parameter explosion. SPAM outperforms all current interpretable approaches, and matches DNN/XGBoost performance on a series of real-world benchmarks with up to hundreds of thousands of features. We demonstrate by human subject evaluations that SPAMs are demonstrably more interpretable in practice, and are hence an effortless replacement for DNNs for creating interpretable and high-performance systems suitable for large-scale machine learning. Source code is available at https://github.com/facebookresearch/nbm-spam.

In general, H is a structured matrix defined in such a way that the product Hu corresponds to a convolution between the image u and a given kernel h representing the Point Spread Function (PSF) of the imaging system employed to measure the data. The deblurring (or deconvolution) problem consists in finding an approximation of g, given the blurred image f and, possibly, some information on the system PSF. If the blurring kernel h, underlying the matrix H, is completely unknown and it has to be inferred together with g, the resulting problem is a blind deconvolution one [32]. Since the PSF h usually represents a low-pass filter, the matrix H is, at best, very ill conditioned and directly solving the inverse problem Hu = f, even when it is feasible, leads to unmeaningful solutions. On the other side, the variational approach consists in designing and solving an optimization problem whose solutions are a good approximation of the unknown image g. In general, a variational model is the set composed by the objective function, i.e., the function to be minimized, and the possible constraints. In the variational models arising in image restoration applications, the objective function, called also energy functional, encompasses different kinds of information: the nature of the noise introduced in the acquisition process, geometrical and/or analytical properties on the image content and physical constraints on the pixel values. Usually, in all image reconstruction problems, and more generally inverse problems, the energy functional, besides the data, depends on a set of parameters; they may simply reduce to tuning parameters balancing the relative weights of the different terms in the functional but can also represent more complicate structures of the functionals themselves.

Even though the large-scale language models have achieved excellent performances, they suffer from various adversarial attacks. A large body of defense methods has been proposed. However, they are still limited due to redundant attack search spaces and the inability to defend against various types of attacks. In this work, we present a novel fine-tuning approach called \textbf{RO}bust \textbf{SE}letive fine-tuning (\textbf{ROSE}) to address this issue. ROSE conducts selective updates when adapting pre-trained models to downstream tasks, filtering out invaluable and unrobust updates of parameters. Specifically, we propose two strategies: the first-order and second-order ROSE for selecting target robust parameters. The experimental results show that ROSE achieves significant improvements in adversarial robustness on various downstream NLP tasks, and the ensemble method even surpasses both variants above. Furthermore, ROSE can be easily incorporated into existing fine-tuning methods to improve their adversarial robustness further. The empirical analysis confirms that ROSE eliminates unrobust spurious updates during fine-tuning, leading to solutions corresponding to flatter and wider optima than the conventional method. Code is available at \url{https://github.com/jiangllan/ROSE}.

A key challenge in the practical application of Gaussian processes (GPs) is selecting a proper covariance function. The moving average, or process convolutions, construction of GPs allows some additional flexibility, but still requires choosing a proper smoothing kernel, which is non-trivial. Previous approaches have built covariance functions by using GP priors over the smoothing kernel, and by extension the covariance, as a way to bypass the need to specify it in advance. However, such models have been limited in several ways: they are restricted to single dimensional inputs, e.g. time; they only allow modelling of single outputs and they do not scale to large datasets since inference is not straightforward. In this paper, we introduce a nonparametric process convolution formulation for GPs that alleviates these weaknesses by using a functional sampling approach based on Matheron's rule to perform fast sampling using interdomain inducing variables. Furthermore, we propose a composition of these nonparametric convolutions that serves as an alternative to classic deep GP models, and allows the covariance functions of the intermediate layers to be inferred from the data. We test the performance of our model on benchmarks for single output GPs, multiple output GPs and deep GPs and find that our approach can provide improvements over standard GP models, particularly for larger datasets.

A fundamental challenge of recommendation systems (RS) is understanding the causal dynamics underlying users' decision making. Most existing literature addresses this problem by using causal structures inferred from domain knowledge. However, there are numerous phenomenons where domain knowledge is insufficient, and the causal mechanisms must be learnt from the feedback data. Discovering the causal mechanism from RS feedback data is both novel and challenging, since RS itself is a source of intervention that can influence both the users' exposure and their willingness to interact. Also for this reason, most existing solutions become inappropriate since they require data collected free from any RS. In this paper, we first formulate the underlying causal mechanism as a causal structural model and describe a general causal structure learning framework grounded in the real-world working mechanism of RS. The essence of our approach is to acknowledge the unknown nature of RS intervention. We then derive the learning objective from our framework and propose an augmented Lagrangian solver for efficient optimization. We conduct both simulation and real-world experiments to demonstrate how our approach compares favorably to existing solutions, together with the empirical analysis from sensitivity and ablation studies.

Predictive Process Monitoring is a field of Process Mining that aims at predicting how an ongoing execution of a business process will develop in the future using past process executions recorded in event logs. The recent stream of publications in this field shows the need for tools able to support researchers and users in analyzing, comparing and selecting the techniques that are the most suitable for them. Nirdizati is a dedicated tool for supporting users in building, comparing, analyzing, and explaining predictive models that can then be used to perform predictions on the future of an ongoing case. By providing a rich set of different state-of-the-art approaches, Nirdizati offers BPM researchers and practitioners a useful and flexible instrument for investigating and comparing Predictive Process Monitoring techniques. In this paper, we present the current version of Nirdizati, together with its architecture which has been developed to improve its modularity and scalability. The features of Nirdizati enrich its capability to support researchers and practitioners within the entire pipeline for constructing reliable Predictive Process Monitoring models.

We propose a Web-Mashup Application Service Framework for Multivariate Time Series Analytics (MTSA) that supports the services of model definitions, querying, parameter learning, model evaluations, data monitoring, decision recommendations, and web portals. This framework maintains the advantage of combining the strengths of both the domain-knowledge-based and the formal-learning-based approaches and is designed for a more general class of problems over multivariate time series. More specifically, we identify a general-hybrid-based model, MTSA-Parameter Estimation, to solve this class of problems in which the objective function is maximized or minimized from the optimal decision parameters regardless of particular time points. This model also allows domain experts to include multiple types of constraints, e.g., global constraints and monitoring constraints. We further extend the MTSA data model and query language to support this class of problems for the services of learning, monitoring, and recommendation. At the end, we conduct an experimental case study for a university campus microgrid as a practical example to demonstrate our proposed framework, models, and language.