The integration of Artificial Intelligence in the development of computer systems presents a new challenge: make intelligent systems explainable to humans. This is especially vital in the field of health and well-being, where transparency in decision support systems enables healthcare professionals to understand and trust automated decisions and predictions. To address this need, tools are required to guide the development of explainable AI systems. In this paper, we introduce an evaluation framework designed to support the development of explainable AI systems for health and well-being. Additionally, we present a case study that illustrates the application of the framework in practice. We believe that our framework can serve as a valuable tool not only for developing explainable AI systems in healthcare but also for any AI system that has a significant impact on individuals.

Traditionally, authorship attribution (AA) tasks relied on statistical data analysis and classification based on stylistic features extracted from texts. In recent years, pre-trained language models (PLMs) have attracted significant attention in text classification tasks. However, although they demonstrate excellent performance on large-scale short-text datasets, their effectiveness remains under-explored for small samples, particularly in AA tasks. Additionally, a key challenge is how to effectively leverage PLMs in conjunction with traditional feature-based methods to advance AA research. In this study, we aimed to significantly improve performance using an integrated integrative ensemble of traditional feature-based and modern PLM-based methods on an AA task in a small sample. For the experiment, we used two corpora of literary works to classify 10 authors each. The results indicate that BERT is effective, even for small-sample AA tasks. Both BERT-based and classifier ensembles outperformed their respective stand-alone models, and the integrated ensemble approach further improved the scores significantly. For the corpus that was not included in the pre-training data, the integrated ensemble improved the F1 score by approximately 14 points, compared to the best-performing single model. Our methodology provides a viable solution for the efficient use of the ever-expanding array of data processing tools in the foreseeable future.

Discovery gene-disease links is important in biology and medicine areas, enabling disease identification and drug repurposing. Machine learning approaches accelerate this process by leveraging biological knowledge represented in ontologies and the structure of knowledge graphs. Still, many existing works overlook ontologies explicitly representing diseases, missing causal and semantic relationships between them. The gene-disease association problem naturally frames itself as a link prediction task, where embedding algorithms directly predict associations by exploring the structure and properties of the knowledge graph. Some works frame it as a node-pair classification task, combining embedding algorithms with traditional machine learning algorithms. This strategy aligns with the logic of a machine learning pipeline. However, the use of negative examples and the lack of validated gene-disease associations to train embedding models may constrain its effectiveness. This work introduces a novel framework for comparing the performance of link prediction versus node-pair classification tasks, analyses the performance of state of the art gene-disease association approaches, and compares the different order-based formalizations of gene-disease association prediction. It also evaluates the impact of the semantic richness through a disease-specific ontology and additional links between ontologies. The framework involves five steps: data splitting, knowledge graph integration, embedding, modeling and prediction, and method evaluation. Results show that enriching the semantic representation of diseases slightly improves performance, while additional links generate a greater impact. Link prediction methods better explore the semantic richness encoded in knowledge graphs. Although node-pair classification methods identify all true positives, link prediction methods outperform overall.

Fairness in artificial intelligence (AI) prediction models is increasingly emphasized to support responsible adoption in high-stakes domains such as health care and criminal justice. Guidelines and implementation frameworks highlight the importance of both predictive accuracy and equitable outcomes. However, current fairness toolkits often evaluate classification performance disparities in isolation, with limited attention to other critical aspects such as calibration. To address these gaps, we present seeBias, an R package for comprehensive evaluation of model fairness and predictive performance. seeBias offers an integrated evaluation across classification, calibration, and other performance domains, providing a more complete view of model behavior. It includes customizable visualizations to support transparent reporting and responsible AI implementation. Using public datasets from criminal justice and healthcare, we demonstrate how seeBias supports fairness evaluations, and uncovers disparities that conventional fairness metrics may overlook. The R package is available on GitHub, and a Python version is under development.

Machine unlearning is a promising approach to improve LLM safety by removing unwanted knowledge from the model. However, prevailing gradient-based unlearning methods suffer from issues such as high computational costs, hyperparameter instability, poor sequential unlearning capability, vulnerability to relearning attacks, low data efficiency, and lack of interpretability. While Sparse Autoencoders are well-suited to improve these aspects by enabling targeted activation-based unlearning, prior approaches underperform gradient-based methods. This work demonstrates that, contrary to these earlier findings, SAEs can significantly improve unlearning when employed dynamically. We introduce $\textbf{Dynamic DAE Guardrails}$ (DSG), a novel method for precision unlearning that leverages principled feature selection and a dynamic classifier. Our experiments show DSG substantially outperforms leading unlearning methods, achieving superior forget-utility trade-offs. DSG addresses key drawbacks of gradient-based approaches for unlearning -- offering enhanced computational efficiency and stability, robust performance in sequential unlearning, stronger resistance to relearning attacks, better data efficiency including zero-shot settings, and more interpretable unlearning.

The performance of algorithmic decision rules is largely dependent on the quality of training datasets available to them. Biases in these datasets can raise economic and ethical concerns due to the resulting algorithms' disparate treatment of different groups. In this paper, we propose algorithms for sequentially debiasing the training dataset through adaptive and bounded exploration in a classification problem with costly and censored feedback. Our proposed algorithms balance between the ultimate goal of mitigating the impacts of data biases -- which will in turn lead to more accurate and fairer decisions, and the exploration risks incurred to achieve this goal. Specifically, we propose adaptive bounds to limit the region of exploration, and leverage intermediate actions which provide noisy label information at a lower cost. We analytically show that such exploration can help debias data in certain distributions, investigate how {algorithmic fairness interventions} can work in conjunction with our proposed algorithms, and validate the performance of these algorithms through numerical experiments on synthetic and real-world data.

Although digital breast tomosynthesis (DBT) improves diagnostic performance over full-field digital mammography (FFDM), false-positive recalls remain a concern in breast cancer screening. We developed a multi-modal artificial intelligence system integrating FFDM, synthetic mammography, and DBT to provide breast-level predictions and bounding-box localizations of suspicious findings. Our AI system, trained on approximately 500,000 mammography exams, achieved 0.945 AUROC on an internal test set. It demonstrated capacity to reduce recalls by 31.7% and radiologist workload by 43.8% while maintaining 100% sensitivity, underscoring its potential to improve clinical workflows. External validation confirmed strong generalizability, reducing the gap to a perfect AUROC by 35.31%-69.14% relative to strong baselines. In prospective deployment across 18 sites, the system reduced recall rates for low-risk cases. An improved version, trained on over 750,000 exams with additional labels, further reduced the gap by 18.86%-56.62% across large external datasets. Overall, these results underscore the importance of utilizing all available imaging modalities, demonstrate the potential for clinical impact, and indicate feasibility of further reduction of the test error with increased training set when using large-capacity neural networks.

Given the demand for responsible and trustworthy AI for education, this study evaluates symbolic, sub-symbolic, and neural-symbolic AI (NSAI) in terms of generalizability and interpretability. Our extensive experiments on balanced and imbalanced self-regulated learning datasets of Estonian primary school students predicting 7th-grade mathematics national test performance showed that symbolic and sub-symbolic methods performed well on balanced data but struggled to identify low performers in imbalanced datasets. Interestingly, symbolic and sub-symbolic methods emphasized different factors in their decision-making: symbolic approaches primarily relied on cognitive and motivational factors, while sub-symbolic methods focused more on cognitive aspects, learnt knowledge, and the demographic variable of gender -- yet both largely overlooked metacognitive factors. The NSAI method, on the other hand, showed advantages by: (i) being more generalizable across both classes -- even in imbalanced datasets -- as its symbolic knowledge component compensated for the underrepresented class; and (ii) relying on a more integrated set of factors in its decision-making, including motivation, (meta)cognition, and learnt knowledge, thus offering a comprehensive and theoretically grounded interpretability framework. These contrasting findings highlight the need for a holistic comparison of AI methods before drawing conclusions based solely on predictive performance. They also underscore the potential of hybrid, human-centred NSAI methods to address the limitations of other AI families and move us closer to responsible AI for education. Specifically, by enabling stakeholders to contribute to AI design, NSAI aligns learned patterns with theoretical constructs, incorporates factors like motivation and metacognition, and strengthens the trustworthiness and responsibility of educational data mining.

In this paper, we seek to develop a versatile test-time adaptation (TTA) objective for a variety of tasks - classification and regression across image-, object-, and pixel-level predictions. We achieve this through a self-bootstrapping scheme that optimizes prediction consistency between the test image (as target) and its deteriorated view. The key challenge lies in devising effective augmentations/deteriorations that: i) preserve the image's geometric information, e.g., object sizes and locations, which is crucial for TTA on object/pixel-level tasks, and ii) provide sufficient learning signals for TTA. To this end, we analyze how common distribution shifts affect the image's information power across spatial frequencies in the Fourier domain, and reveal that low-frequency components carry high power and masking these components supplies more learning signals, while masking high-frequency components can not. In light of this, we randomly mask the low-frequency amplitude of an image in its Fourier domain for augmentation. Meanwhile, we also augment the image with noise injection to compensate for missing learning signals at high frequencies, by enhancing the information power there. Experiments show that, either independently or as a plug-and-play module, our method achieves superior results across classification, segmentation, and 3D monocular detection tasks with both transformer and CNN models.

Machine learning models are routinely used to support decisions that affect individuals - be it to screen a patient for a serious illness or to gauge their response to treatment. In these tasks, we are limited to learning models from datasets with noisy labels. In this paper, we study the instance-level impact of learning under label noise. We introduce a notion of regret for this regime which measures the number of unforeseen mistakes due to noisy labels. We show that standard approaches to learning under label noise can return models that perform well at a population level while subjecting individuals to a lottery of mistakes . We present a versatile approach to estimate the likelihood of mistakes at the individual level from a noisy dataset by training models over plausible realizations of datasets without label noise. This is supported by a comprehensive empirical study of label noise in clinical prediction tasks. Our results reveal how failure to anticipate mistakes can compromise model reliability and adoption, and demonstrate how we can address these challenges by anticipating and avoiding regretful decisions. Machine learning models are routinely used to support or automate decisions that affect individuals - be it to screen a patient for a mental illness [47], or assess their risk for an adverse treatment response [3]. In such tasks, we train models with labels that reflect noisy observations of the true outcome we wish to predict. In practice, such noise may arise due to measurement error [e.g., 20, 35], human annotation [26], or inherent ambiguity [35]. In all these cases, label noise can have detrimental effects on model performance [10]. Over the past decade, these issues have led to extensive work on learning from noisy datasets [see e.g., 10, 28, 36, 39, 45].