Segment Anything Model 2 (SAM2) has demonstrated state-of-the-art performance in image/video object segmentation across many domains, but its large encoder makes it challenging for resource-constrained devices or real-time applications. One solution to this problem is to carry out knowledge distillation from the bulky encoder to a lightweight encoder, but this can result in degraded performance. In this work, we investigate multi-teacher distillation to mitigate performance degradation for distilled segmentation models. Using several foundation teacher models, our multi-teacher distilled models achieve 3.2 times speedup during end-to-end inference compared to SAM2 while achieving the best results of 74.4 and 71.1 (72.1 and 69.6 for single-teacher distillation) mIoU on the COCO and LVIS image segmentation datasets, as well as showing competitive results on video segmentation. Our results show that multi-teacher distillation offers a powerful solution for efficient image/video segmentation, while also maintaining compelling performance.

In supervised learning, automatically assessing the quality of the labels before any learning takes place remains an open research question. In certain particular cases, hypothesis testing procedures have been proposed to assess whether a given instance-label dataset is contaminated with class-conditional label noise, as opposed to uniform label noise. The existing theory builds on the asymptotic properties of the Maximum Likelihood Estimate for parametric logistic regression. However, the parametric assumptions on top of which these approaches are constructed are often too strong and unrealistic in practice. To alleviate this problem, in this paper we propose an alternative path by showing how similar procedures can be followed when the underlying model is a product of Local Maximum Likelihood Estimation that leads to more flexible nonparametric logistic regression models, which in turn are less susceptible to model misspecification. This different view allows for wider applicability of the tests by offering users access to a richer model class. Similarly to existing works, we assume we have access to anchor points which are provided by the users. We introduce the necessary ingredients for the adaptation of the hypothesis tests to the case of nonparametric logistic regression and empirically compare against the parametric approach presenting both synthetic and real-world case studies and discussing the advantages and limitations of the proposed approach.

Conventional Time Series Classification (TSC) methods are often black boxes that obscure inherent interpretation of their decision-making processes. In this work, we leverage Multiple Instance Learning (MIL) to overcome this issue, and propose a new framework called MILLET: Multiple Instance Learning for Locally Explainable Time series classification. We apply MILLET to existing deep learning TSC models and show how they become inherently interpretable without compromising (and in some cases, even improving) predictive performance. We evaluate MILLET on 85 UCR TSC datasets and also present a novel synthetic dataset that is specially designed to facilitate interpretability evaluation. On these datasets, we show MILLET produces sparse explanations quickly that are of higher quality than other well-known interpretability methods. To the best of our knowledge, our work with MILLET is the first to develop general MIL methods for TSC and apply them to an extensive variety of domains.

Cross-lingual text representations have gained popularity lately and act as the backbone of many tasks such as unsupervised machine translation and cross-lingual information retrieval, to name a few. However, evaluation of such representations is difficult in the domains beyond standard benchmarks due to the necessity of obtaining domain-specific parallel language data across different pairs of languages. In this paper, we propose an automatic metric for evaluating the quality of cross-lingual textual representations using images as a proxy in a paired image-text evaluation dataset. Experimentally, Backretrieval is shown to highly correlate with ground truth metrics on annotated datasets, and our analysis shows statistically significant improvements over baselines. Our experiments conclude with a case study on a recipe dataset without parallel cross-lingual data. We illustrate how to judge cross-lingual embedding quality with Backretrieval, and validate the outcome with a small human study.

It was recently shown that neural ordinary differential equation models cannot solve fundamental and seemingly straightforward tasks even with high-capacity vector field representations. This paper introduces two other fundamental tasks to the set that baseline methods cannot solve, and proposes mixtures of stochastic vector fields as a model class that is capable of solving these essential problems. Dynamic vector field selection is of critical importance for our model, and our approach is to propagate component uncertainty over the integration interval with a technique based on forward filtering. We also formalise several loss functions that encourage desirable properties on the trajectory paths, and of particular interest are those that directly encourage fewer expected function evaluations. Experimentally, we demonstrate that our model class is capable of capturing the natural dynamics of human behaviour; a notoriously volatile application area. Baseline approaches cannot model this problem.

Multi-language recipe personalisation and recommendation is an under-explored field of information retrieval in academic and production systems. The existing gaps in our current understanding are numerous, even on fundamental questions such as whether consistent and high-quality recipe recommendation can be delivered across languages. Motivated by this need, we consider the multi-language recipe recommendation setting and present grounding results that will help to establish the potential and absolute value of future work in this area. Our work draws on several billion events from millions of recipes, with published recipes and users incorporating several languages, including Arabic, English, Indonesian, Russian, and Spanish. We represent recipes using a combination of normalised ingredients, standardised skills and image embeddings obtained without human intervention. In modelling, we take a classical approach based on optimising an embedded bi-linear user-item metric space towards the interactions that most strongly elicit cooking intent. For users without interaction histories, a bespoke content-based cold-start model that predicts context and recipe affinity is introduced. We show that our approach to personalisation is stable and scales well to new languages. A robust cross-validation campaign is employed and consistently rejects baseline models and representations, strongly favouring those we propose. Our results are presented in a language-oriented (as opposed to model-oriented) fashion to emphasise the language-based goals of this work. We believe that this is the first large-scale work that evaluates the value and potential of multi-language recipe recommendation and personalisation.

Active Learning (AL) refers to the setting where the learner has the ability to perform queries to an oracle to acquire the true label of an instance or, sometimes, a set of instances. Even though Active Learning has been studied extensively, the setting is usually restricted to assume that the oracle is trustworthy and will provide the actual label. We argue that, while common, this approach can be made more flexible to account for different forms of supervision. In this paper, we propose a new framework that allows the algorithm to request the label for a bag of samples at a time. Although this label will come in the form of proportions of class labels in the bags and therefore encode less information, we demonstrate that we can still learn effectively.

The SPHERE project is devoted to advancing eHealth in a smart-home context, and supports full-scale sensing and data analysis to enable a generic healthcare service. We describe, from a data-science perspective, our experience of taking the system out of the laboratory into more than thirty homes in Bristol, UK. We describe the infrastructure and processes that had to be developed along the way, describe how we train and deploy Machine Learning systems in this context, and give a realistic appraisal of the state of the deployed systems.

There is no uniform approach in the literature for modelling sequential correlations in sequence classification problems. It is easy to find examples of unstructured models ( e.g. logistic regression) where correlations are not taken into account at all, but there are also many examples where the correlations are explicitly incorporated into a—potentially computationally expensive—structured classification model ( e.g. conditional random fields). In this paper we lay theoretical and empirical foundations for clarifying the types of problem which necessitate direct modelling of correlations in sequences, and the types of problem where unstructured models that capture sequential aspects solely through features are sufficient. The theoretical work in this paper shows that the rate of decay of auto-correlations within a sequence is related to the excess classification risk that is incurred by ignoring the structural aspect of the data. This is an intuitively appealing result, demonstrating the intimate link between the auto-correlations and excess classification risk. Drawing directly on this theory, we develop well-founded visual analytics tools that can be applied a priori on data sequences and we demonstrate how these tools can guide practitioners in specifying feature representations based on auto-correlation profiles. Empirical analysis is performed on three sequential datasets. With baseline feature templates, structured and unstructured models achieve similar performance, indicating no initial preference for either model. We then apply the visual analytics tools to the datasets, and show that classification performance in all cases is improved over baseline results when our tools are involved in defining feature representations.

This paper presents an automated framework for the detection of the QRS complex from Electrocardiogram (ECG) signals. We introduce an artefact-tolerant pre-processing algorithm which emphasises a number of characteristics of the ECG that are representative of the QRS complex. With this processed ECG signal we train Logistic Regression and Support Vector Machine classification models. With our approach we obtain over 99.7% detection sensitivity and precision on the MIT-BIH database without using supplementary de-noising or pre-emphasis filters.

This paper investigates the fully automated computer-based detection of allergic reaction in oral food challenges using pediatric ECG signals. Nonallergic background is modeled using a mixture of Gaussians during oral food challenges, and the model likelihoods are used to determine whether a subject is allergic to a food type. The system performance is assessed on the dataset of 24 children (15 allergic and 9 nonallergic) totaling 34 h of data. The proposed detector correctly classified all nonallergic subjects (100% specificity) and 12 allergic subjects (80% sensitivity) and is capable of detecting allergy on average 17 min earlier than trained clinicians during oral food challenges, the gold standard of allergy diagnosis. Inclusion of the developed allergy classification platform during oral food challenges recorded would result in a 30% reduction of doses administered to allergic subjects. The results of study introduce the possibility to halt challenges earlier which can safely advance the state of clinical art of allergy diagnosis by reducing the overall exposure to the allergens.