Part of the success of an online media platform depends on its ability to convert first-time users into recurring ones. However, this task is often challenging due to the Pure Cold-start problem, which refers to the difficulty of providing useful recommendations to users without historical data. Although presenting a list of popular items may seem like an easy fix, it can lead to the ”popularity bias” problem. In this article, we address a specific issue: the Contextual Pure Cold-start Problem in Public Service Media (PSM) scenarios, characterized by the absence of user-specific data and the presence of only anonymous and limited contextual information. We propose a novel approach aligned with PSM values in recommendations, as ensuring these values is crucial for their task. Our approach seeks to enhance the equity of recommendation rankings by incorporating various types of uncertainty, providing a solution that ensures a fairer exposure to items and broader coverage of the PSM catalog. Using a substantial dataset of real interactions from a public television network, our approach successfully mitigates the ”popularity bias” issue through the use of an uncertainty-based stochastic ranker. Consequently, we achieve a 64% improvement in fair exposure and a 42% increase in coverage metrics, with only an 8% reduction in Hit Rate accuracy metrics.

AI becomes increasingly important in safetycritical systems, especially in the case of autonomous systems, since navigation relies on AI for object detection and collision avoidance. However, safety-critical systems must adhere to functional safety standards that enforce software to be correct-by-construction, component decomposition to simplify design and validation, and the use of data only for testing purposes not to design the system itself. AI in general, and Deep Learning (DL) in particular have opposed characteristics since they have error rates (e.g., due to mispredictions), AI/DL modules can only be designed and validated monolithically, and they build on data for their design (i.e. for training purposes). Hence, DL solutions are at odds with the development process of safetycritical systems. A number of standards have recently emerged in different domains to reconcile the requirements of safety-critical systems with the characteristics of DL solutions, such as ISO 21448, ISO/IEC TR 5469, and ISO 8800, among others. However, there is a lack of realistic practice to design a DL-based safety-critical system in accordance with those regulations, and existing solutions only cover some aspects in isolation, and are often incompatible among them. SAFEXPLAIN is a 3-year Horizon Europe project addressing this challenge. SAFEXPLAIN, which finishes in September 2025, has already reached its main goals providing specific and complementary solutions to all those challenges so that AIbased safety-critical systems can be designed, implemented and validated adhering to the relevant functional safety standards in domains such as automotive, space and railway. In particular, SAFEXPLAIN provides the concepts, processes, tools and frameworks addressing the challenge end-to-end, from concept to solution. This is proven by the successful application of the SAFEXPLAIN approach in three case studies from the automotive, space and railway domains, whose results will see the light very soon.

Enhancing the robustness of object detection systems under adverse weather conditions is crucial for the advancement of autonomous driving technology. This study presents a novel approach leveraging the diffusion model Instruct Pix2Pix to develop prompting methodologies that generate realistic datasets with weather-based augmentations aiming to mitigate the impact of adverse weather on the perception capabilities of state-of-the-art object detection models, including Faster R-CNN and YOLOv10. Experiments were conducted in two environments, in the CARLA simulator where an initial evaluation of the proposed data augmentation was provided, and then on the real-world image data sets BDD100K and ACDC demonstrating the effectiveness of the approach in real environments.The key contributions of this work are twofold: (1) identifying and quantifying the performance gap in object detection models under challenging weather conditions, and (2) demonstrating how tailored data augmentation strategies can significantly enhance the robustness of these models. This research establishes a solid foundation for improving the reliability of perception systems in demanding environmental scenarios, and provides a pathway for future advancements in autonomous driving.

Dual (DMR) and Triple Modular Redundancy (TMR) are widely used techniques to provide fault detection and/or tolerance capabilities in safety-critical systems through – often diverse – redundancy. However, these systems remain vulnerable to adversarial attacks, which can mislead the AI models and lead to severe consequences. In this paper, we propose enhanced DMR and TMR implementations for image-based object detection leveraging image transformations during inference to mitigate the impact of adversarial attacks, hence addressing safety and security concerns simultaneously. Our approach achieves up to 12.9% and 12.2% higher accuracy in adversarial scenarios compared to state-of-the-art solutions in DMR and TMR configurations, respectively.

Artificial intelligence (AI) now permeates critical infrastructures and decisionmaking systems where failures produce social, economic, and democratic harm. This position paper challenges the entrenched belief that regulation and innovation are opposites. As evidenced by analogies from aviation, pharmaceuticals, and welfare systems and recent cases of synthetic misinformation, bias and unaccountable decision-making, the absence of well-designed regulation has already created immeasurable damage. Regulation, when thoughtful and adaptive, is not a brake on innovation—it is its foundation. The present position paper examines the EU AI Act as a model of risk-based, responsibility-driven regulation that addresses the Collingridge Dilemma: acting early enough to prevent harm, yet flexibly enough to sustain innovation. Its adaptive mechanisms—regulatory sandboxes, small and medium enterprises (SMEs) support, real-world testing, fundamental rights impact assessment (FRIA)—demonstrate how regulation can accelerate responsibly, rather than delay, technological progress. The position paper summarises how governance tools transform perceived burdens into tangible advantages: legal certainty, consumer trust, and ethical competitiveness. Ultimately, the paper reframes progress: innovation and regulation advance together. By embedding transparency, impact assessments, accountability, and AI literacy into design and deployment, the EU framework defines what responsible innovation truly means—technological ambition disciplined by democratic values and fundamental rights.

Among explainability techniques, SHAP stands out as one of the most popular, but often overlooks the causal structure of the problem. In response, do-SHAP employs interventional queries, but its reliance on estimands hinders its practical application. To address this problem, we propose the use of estimand-agnostic approaches, which allow for the estimation of any identifiable query from a single model, making do-SHAP feasible on complex graphs. We also develop a novel algorithm to significantly accelerate its computation at a negligible cost, as well as a method to explain inaccessible Data Generating Processes. We demonstrate the estimation and computational performance of our approach, and validate it on two real-world datasets, highlighting its potential in obtaining reliable explanations.

Multicollinearity of Event Monitors (EMs) negatively impacts the modeling of non-functional critical metrics in real-time systems like worst-case timing and energy usage since some EMs are over-represented and can reduce model accuracy. To address this challenge, we propose Event Monitor Reduction (EMR), a method to select a reduced set of non-related (independent) features (EMs), hence eliminating multicollinearity. In particular, EMR finds linear relations between the EMs and removes dependent ones without data loss. EMR does not create new features like Principal Component Analysis does, simplifying interpretability. Results on synthetic data and data collected from the execution of representative benchmarks on an avionicsgrade processor show the benefits of our method in removing multicollinear EMs. We further illustrate the benefits of EMR on two different multicore timing contention models, showing how its application helps to reduce execution time requirements and increase the accuracy of the models.

Dual Modular Redundancy (DMR) and Triple Modular Redundancy (TMR), often with some form of diversity, are used in safety-critical systems to realize those functionalities at the highest integrity level providing fault detection and/or tolerance capabilities. Redundant executions are intended to provide bit-level identical results, and, upon any mismatch, an error is assumed and recovery actions taken as needed. In this article, we note that many emerging AI-based functionalities are intrinsically stochastic (e.g., camera-based object detection), and hence, their correctness must be judged semantically, with room for variations across correct outcomes (e.g., confidence must be above a given threshold, but how much it exceeds the threshold is irrelevant). Building on this observation, we propose strategies to create DMR and TMR implementations of AI-based functionalities that bring not only fault tolerance against random hardware faults but also against AI model inaccuracies. Those strategies, which can be realized with software-only means and ported to virtually any computing platform, build on input data modifications affecting the inference computations, but not the expected semantic output (e.g., introducing some controlled changes in the input data). Moreover, we provide our solution in the form of an open source tool for image and video processing aimed at facilitating the reproducibility of our evaluation results, and enabling others to use it and conduct further research on input transformations.

Functional safety requirements of AI-based safety critical applications challenge AI models, whose accuracy can be limited. In this paper, we show how using several cooperative deep learning (DL) models helps to raise global accuracy and reject making a prediction when confidence is below a pre-established threshold.

Generative diffusion models, notable for their large parameter count (exceeding 100 million) and operation within high-dimensional image spaces, pose significant challenges for traditional uncertainty estimation methods due to computational demands. In this work, we introduce an innovative framework, Diffusion Ensembles for Capturing Uncertainty (DECU), designed for estimating epistemic uncertainty for diffusion models. The DECU framework introduces a novel method that efficiently trains ensembles of conditional diffusion models by incorporating a static set of pre-trained parameters, drastically reducing the computational burden and the number of parameters that require training. Additionally, DECU employs Pairwise-Distance Estimators (PaiDEs) to accurately measure epistemic uncertainty by evaluating the mutual information between model outputs and weights in high-dimensional spaces. The effectiveness of this framework is demonstrated through experiments on the ImageNet dataset, highlighting its capability to capture epistemic uncertainty, specifically in under-sampled image classes.

Despite the major progress of deep models as learning machines, uncertainty estimation remains a major challenge. Existing solutions rely on modified loss functions or architectural changes. We propose to compensate for the lack of built-in uncertainty estimates by supplementing any network, retrospectively, with a subsequent vine copula model, in an overall compound we call Vine-Copula Neural Network (VCNN). Through synthetic and real-data experiments, we show that VCNNs could be task (regression/classification) and architecture (recurrent, fully connected) agnostic while providing reliable and better-calibrated uncertainty estimates, comparable to state-of-the-art built-in uncertainty solutions.

Deep Learning (DL) techniques are at the heart of most future advanced software functions in Critical Autonomous AI-based Systems (CAIS), where they also represent a major competitive factor. Hence, the economic success of CAIS industries (e.g., automotive, space, railway) depends on their ability to design, implement, qualify, and certify DL-based software products under bounded effort/cost. However, there is a fundamental gap between Functional Safety (FUSA) requirements on CAIS and the nature of DL solutions. This gap stems from the development process of DL libraries and affects high-level safety concepts such as (1) explainability and traceability, (2) suitability for varying safety requirements, (3) FUSA-compliant implementations, and (4) real-time constraints. As a matter of fact, the data-dependent and stochastic nature of DL algorithms clashes with current FUSA practice, which instead builds on deterministic, verifiable, and pass/fail test-based software. The SAFEXPLAIN project tackles these challenges and targets by providing a flexible approach to allow the certification - hence adoption - of DL-based solutions in CAIS building on: (1) DL solutions that provide end-to-end traceability, with specific approaches to explain whether predictions can be trusted and strategies to reach (and prove) correct operation, in accordance to certification standards; (2) alternative and increasingly sophisticated design safety patterns for DL with varying criticality and fault tolerance requirements; (3) DL library implementations that adhere to safety requirements; and (4) computing platform configurations, to regain determinism, and probabilistic timing analyses, to handle the remaining non-determinism.

This special session paper introduces the Horizon Europe NEUROPULS project, which targets the development of secure and energy-efficient RISC-V interfaced neuromorphic accelerators using augmented silicon photonics technology. Our approach aims to develop an augmented silicon photonics platform, an FPGA-powered RISC-V-connected computing platform, and a complete simulation platform to demonstrate the neuromorphic accelerator capabilities. In particular, their main advantages and limitations will be addressed concerning the underpinning technology for each platform. Then, we will discuss three targeted use cases for edge-computing applications: Global National Satellite System (GNSS) anti-jamming, autonomous driving, and anomaly detection in edge devices. Finally, we will address the reliability and security aspects of the stand-alone accelerator implementation and the project use cases.

Quantile Regression (QR) provides a way to approximate a single conditional quantile. To have a more informative description of the conditional distribution, QR can be merged with deep learning techniques to simultaneously estimate multiple quantiles. However, the minimisation of the QR-loss function does not guarantee non-crossing quantiles, which affects the validity of such predictions and introduces a critical issue in certain scenarios. In this article, we propose a generic deep learning algorithm for predicting an arbitrary number of quantiles that ensures the quantile monotonicity constraint up to the machine precision and maintains its modelling performance with respect to alternative models. The presented method is evaluated over several real-world datasets obtaining state-of-the-art results as well as showing that it scales to large-size data sets.

Machine learning has enabled significant benefits in diverse fields, but, with a few exceptions, has had limited impact on computer architecture. Recent work, however, has explored broader applicability for design, optimization, and simulation. Notably, machine learning based strategies often surpass prior state-of-the-art analytical, heuristic, and human-expert approaches. This paper reviews machine learning applied system-wide to simulation and run-time optimization, and in many individual components, including memory systems, branch predictors, networks-on-chip, and GPUs. The paper further analyzes current practice to highlight useful design strategies and identify areas for future work, based on optimized implementation strategies, opportune extensions to existing work, and ambitious long term possibilities. Taken together, these strategies and techniques present a promising future for increasingly automated architectural design.

In regression tasks, aleatoric uncertainty is commonly addressed by considering a parametric distribution of the output variable, which is based on strong assumptions such as symmetry, unimodality or by supposing a restricted shape. These assumptions are too limited in scenarios where complex shapes, strong skews or multiple modes are present. In this paper, we propose a generic deep learning framework that learns an Uncountable Mixture of Asymmetric Laplacians (UMAL), which will allow us to estimate heterogeneous distributions of the output variable and shows its connections to quantile regression. Despite having a fixed number of parameters, the model can be interpreted as an infinite mixture of components, which yields a flexible approximation for heterogeneous distributions. Apart from synthetic cases, we apply this model to room price forecasting and to predict financial operations in personal bank accounts. We demonstrate that UMAL produces proper distributions, which allows us to extract richer insights and to sharpen decision-making.