Cerebral Palsy (CP) is a prevalent motor disability in children, for which early detection can significantly improve treatment outcomes. While skeleton-based Graph Convolutional Network (GCN) models have shown promise in automatically predicting CP risk from infant videos, their “black-box” nature raises concerns about clinical explainability. To address this, we introduce a perturbation framework tailored for infant movement features and use it to compare two explainable AI (XAI) methods: Class Activation Mapping (CAM) and Gradient-weighted Class Activation Mapping (Grad-CAM). First, we identify significant and non-significant body keypoints in very low and very high risk infant video snippets based on the XAI attribution scores. We then conduct targeted velocity and angular perturbations, both individually and in combination, on these keypoints to assess how the GCN model’s risk predictions change. Our results indicate that velocity-driven features of the arms, hips, and legs appear to have a dominant influence on CP risk predictions, while angular perturbations have a more modest impact. Furthermore, CAM and Grad-CAM show partial convergence in their explanations for both low and high CP risk groups. Our findings demonstrate the use of XAI-driven movement analysis for early CP prediction, and offer insights into potential movement-based biomarker discovery that warrant further clinical validation.

This work presents a conceptual framework for causal concept-based post-hoc explainable artificial intelligence (XAI), based on the requirements that explanations for non-interpretable models must be both understandable and faithful to the model being explained. Local and global explanations are generated by calculating the probability of sufficiency of concept interventions. Example explanations are presented, generated with a proof-of-concept model made to explain classifiers trained on the CelebA dataset. Understandability is demonstrated through a clear concept-based vocabulary, subject to an implicit causal interpretation. Fidelity is addressed by highlighting important framework assumptions, stressing that the context of explanation interpretation must align with the context of explanation generation.

Cooperative game theorists propose the following attractive process: (1) capture the abstract value of each possible coalition of individuals, (2) write down some principles, or axioms, on how to distribute the value, and then, (3) find a set of allocations that satisfy the principles. The Shapley value has received much attention – but it is just one solution concept, satisfying one set of principles, in one class of games. In our work, we contrast the Shapley value with solution concepts in the class of multi-choice games (MC-games), which extends the class of transferrable-utility games to cases where features or parameters have levels of participation. These MC-game solutions are model agnostic, and unique to their own set of axioms, just like the Shapley value. We demonstrate in a simple setting in the context of polynomial regression that three MC-game solution concepts can outperform the Shapley value. Our work offers a general algorithm for constructing any MC-game framework with polynomial time complexity in the number of parameter levels, and an application of this algorithm that is transparent, and can be used to generalise local explanation frameworks such as SHapley Additive exPlanations (SHAP).

We present swarm-attack, an open-source adversarial testing framework in which multiple lightweight LLM agents coordinate through shared memory, parallel exploration, and evolutionary optimization. Together, our results demonstrate that both safety bypass of frontier models and software vulnerability discovery, i.e., the capability class that motivated restricted release of Anthropic's Mythos Preview, are achievable at effectively zero cost using commodity hardware and openly available models. We report two experiments. In the first, five instances of a 1.2 billion parameter model conducted 225 jailbreak attacks each against GPT-4o and Claude Sonnet 4. Against GPT-4o, the swarm achieved an Effective Harm Rate of 45.8

Recent cryptographic results establish that neural networks can be backdoored such that no efficient algorithm can distinguish them from a clean model. These guarantees, however, have been confined to stylised architectures of limited practical relevance, leaving open whether comparable undetectability extends to modern, end-to-end trained networks. We construct such an attack mechanism for state-of-the-art architectures, closely aligned to the cryptographic notion of undetectability, by identifying backdoor channels as learned latent directions, and show that the question of undetectability reduces to a hypothesis test between two unknown distributions over model parameters, which we conjecture to be intractable in practice. The consequence of this reframing is significant: if exploitable channels within a network's latent space are statistically indistinguishable from naturally learned directions, an attacker need not introduce foreign structure but can instead exploit the geometry the network already possesses. Demonstrating the approach on ResNet and Vision Transformer architectures trained on standard image classification datasets, the attack achieves both consistently high success rates with negligible clean accuracy degradation, and resists a comprehensive suite of post-training defences, none of which neutralise the backdoor without rendering the model unusable. Our results establish that cryptographic backdoors need not be artefacts requiring exotic architectures or artificial constructions, but identifiable as latent properties inherent to the geometry of learned representations.

Today, the use of Artificial Intelligence (AI) is rapidly increasing in many areas of society. While model performance on various tasks continue to impress, it does so at the cost of increased model complexity, such that most state-of-the-art AI models are effectively black boxes. Where human-made decisions typically are accompanied by human-understandable explanations detailing the reasoning behind the decision, incorporating advanced AI as part of a decision-making process reduces the transparency of that process significantly. Yet, the ability to explain decisions is essential for there to be understanding and trust. As a response to this, Explainable Artificial Intelligence (XAI) has emerged as a field that aims to provide explanations of model behaviour. Methods categorised as post-hoc are designed to generate explanations for black box models after training, at no cost to model performance. In parallel with this, extensive work has been done in the field of causality to formalise the structure of human-understandable, causal explanations. This work presents a comprehensive literature review of the current state of the subfield of XAI that consist of causality-motivated post-hoc XAI methods. In order to clearly define causal XAI, a causal framework for categorising XAI is introduced, and three types of post-hoc XAI methods are identified: observational methods, internally causal methods and externally causal methods. Finally, externally causal XAI is argued a promising direction for reliable and understandable post-hoc XAI, with the ability to generate counterfactual explanations using a meaningful vocabulary, in line with the definition of counterfactual used in causal theory.

The neurological condition known as cerebral palsy (CP) first manifests in infancy or early childhood and has a lifelong impact on motor coordination and body movement. CP is one of the leading causes of childhood disabilities, and early detection is crucial for providing appropriate treatment. However, such detection relies on assessments by human experts trained in methods like general movement assessment (GMA). These are not widely accessible, especially in developing countries. Conventional machine learning approaches offer limited predictive performance on CP detection tasks, and the approaches developed by the few available domain experts are generally dataset-specific, restricting their applicability beyond the context for which these were created. To address these challenges, we propose a neural architecture search (NAS) algorithm applying a reinforcement learning update scheme capable of efficiently optimizing for the best architectural and hyperparameter combination to discover the most suitable neural network configuration for detecting CP. Our method performs better on a real-world CP dataset than other approaches in the field, which rely on large ensembles. As our approach is less resource-demanding and performs better, it is particularly suitable for implementation in resourceconstrained settings, including rural or developing areas with limited access to medical experts and the required diagnostic tools. The resulting model's lightweight architecture and efficient computation time allow for deployment on devices with limited processing power, reducing the need for expensive infrastructure, and can, therefore, be integrated into clinical workflows to provide timely and accurate support for early CP diagnosis. Finally, our proposed lightweight approach simplifies the application of computationally intensive explainability methods, as it avoids the scenario where a full ensemble must be explained.

Explaining machine learning (ML) models using eXplainable AI (XAI) techniques has become essential to make them more transparent and trustworthy. This is especially important in high-risk environments like healthcare, where understanding model decisions is critical to ensure ethical, sound, and trustworthy outcome predictions. However, users are often confused about which explanability method to choose for their specific use case. We present a comparative analysis of two explainability methods, Shapley Additive Explanations (SHAP) and Gradient-weighted Class Activation Mapping (Grad-CAM), within the domain of human activity recognition (HAR) utilizing graph convolutional networks (GCNs). By evaluating these methods on skeleton-based input representation from two real-world datasets, including a healthcare-critical cerebral palsy (CP) case, this study provides vital insights into both approaches’ strengths, limitations, and differences, offering a roadmap for selecting the most appropriate explanation method based on specific models and applications. We qualitatively and quantitatively compare the two methods, focusing on feature importance ranking and model sensitivity through perturbation experiments. While SHAP provides detailed input feature attribution, Grad-CAM delivers faster, spatially oriented explanations, making both methods complementary depending on the application’s requirements. Given the importance of XAI in enhancing trust and transparency in ML models, particularly in sensitive environments like healthcare, our research demonstrates how SHAP and Grad-CAM could complement each other to provide model explanations.

Early detection of Cerebral Palsy (CP) is crucial for effective intervention and monitoring. This paper tests the reliability and applicability of Explainable AI (XAI) methods using a deep learning method that predicts CP by analyzing skeletal data extracted from video recordings of infant movements. Specifically, we use XAI evaluation metrics — namely faithfulness and stability — to quantitatively assess the reliability of Class Activation Mapping (CAM) and Gradient-weighted Class Activation Mapping (Grad-CAM) in this specific medical application. We utilize a unique dataset of infant movements and apply skeleton data perturbations without distorting the original dynamics of the infant movements. Our CP prediction model utilizes an ensemble approach, so we evaluate the XAI metrics performances for both the overall ensemble and the individual models. Our findings indicate that both XAI methods effectively identify key body points influencing CP predictions and that the explanations are robust against minor data perturbations. Grad-CAM significantly outperforms CAM in the Relative Input Stability velocity (RISv) metric, which measures stability in terms of velocity. In contrast, CAM performs better in the Relative Input Stability bone (RISb) metric, which relates to bone stability, and the Relative Representation Stability (RRS) metric, which assesses internal representation robustness. Individual models within the ensemble show varied results, and neither CAM nor Grad-CAM consistently outperform the other, with the ensemble approach providing a representation of outcomes from its constituent models. Both CAM and Grad-CAM also perform significantly better than random attribution, supporting the robustness of these XAI methods. Ourwork demonstrates that XAI methods can offer reliable and stable explanations for CP prediction models. Future studies should further investigate how the explanations can enhance our understanding of specific movement patterns characterizing healthy and pathological development.

Despite the advanced predictive capabilities of artificial intelligence (AI) systems, their inherent opacity often leaves users confused about the rationale behind their outputs. We investigate the challenge of AI opacity, which undermines user trust and the effectiveness of clinical judgment in healthcare. We demonstrate how human experts make judgment in high-stakes scenarios where their judgment diverges from AI predictions, emphasizing the need for explainability to enhance clinical judgment and trust in AI systems. We used a scenario-based methodology, conducting 28 semi-structured interviews and observations with clinicians from Norway and Egypt. Our analysis revealed that, during the process of forming judgments, human experts engage in AI interrogation practices when faced with opaque AI systems. Obtaining explainability from AI systems leads to increased interrogation practices aimed at gaining a deeper understanding of AI predictions. With the introduction of explainable AI (XAI), experts demonstrate greater trust in the AI system, show a readiness to learn from AI, and may reconsider or update their initial judgments when they contradict AI predictions.

Navigation is a fundamental cognitive skill extensively studied in neuroscientific experiments and has lately gained substantial interest in artificial intelligence research. Recreating the task solved by rodents in the well-established Morris Water Maze (MWM) experiment, this work applies a transformer-based architecture using deep reinforcement learning - an approach previously unexplored in this context - to navigate a 2D version of the maze. Specifically, the agent leverages a decoder-only transformer architecture serving as a deep Q-network performing effective decision making in the partially observable environment. We demonstrate that the proposed architecture enables the agent to efficiently learn spatial navigation strategies, overcoming challenges associated with a limited field of vision, corresponding to the visual information available to a rodent in the MWM. Demonstrating the potential of transformer-based models for enhancing navigation performance in partially observable environments, this work suggests promising avenues for future research in artificial agents whose behavior resembles that of biological agents. Finally, the flexibility of the transformer architecture in supporting varying input sequence lengths opens opportunities for gaining increased understanding of the artificial agent's inner representation of the environment.

This paper reports a case study on how explainability requirements were elicited during the development of an AI system for predicting cerebral palsy (CP) risk in infants. Over 18 months, we followed a development team and hospital clinicians as they sought to design explanations that would make the AI system trustworthy. Contrary to the assumption that users need detailed explanations of the inner workings of AI systems, our findings show that clinicians trusted it when it enabled them to evaluate predictions against their own assessments. Our findings show how a simple prediction graph proved effective by supporting clinicians' existing decision-making practices. Drawing on concepts from both Requirements Engineering and Explainable AI, we use the theoretical lens of Evaluative AI to introduce the notion of Evaluative Requirements: system requirements that allow users to scrutinize AI outputs and compare them with their own assessments. Our study demonstrates that such requirements are best discovered through the well-known methods of iterative prototyping and observation, making them essential for building trustworthy AI systems in expert domains.

In Human Activity Recognition (HAR), understanding the intricacy of body movements within high-risk applications is essential. This study uses SHapley Additive exPlanations (SHAP) to explain the decision-making process of Graph Convolution Networks (GCNs) when classifying activities with skeleton data. We employ SHAP to explain two real-world datasets: one for cerebral palsy (CP) classification and the widely used NTU RGB+D 60 action recognition dataset. To test the explanation, we introduce a novel perturbation approach that modifies the model's edge importance matrix, allowing us to evaluate the impact of specific body key points on prediction outcomes. To assess the fidelity of our explanations, we employ informed perturbation, targeting body key points identified as important by SHAP and comparing them against random perturbation as a control condition. This perturbation enables a judgment on whether the body key points are truly influential or non-influential based on the SHAP values. Results on both datasets show that body key points identified as important through SHAP have the largest influence on the accuracy, specificity, and sensitivity metrics. Our findings highlight that SHAP can provide granular insights into the input feature contribution to the prediction outcome of GCNs in HAR tasks. This demonstrates the potential for more interpretable and trustworthy models in high-stakes applications like healthcare or rehabilitation.

Deep learning has achieved immense success in computer vision and has the potential to help physicians analyze visual content for disease and other abnormalities. However, the current state of deep learning is very much a black box, making medical professionals skeptical about integrating these methods into clinical practice. Several methods have been proposed to shed some light on these black boxes, but there is no consensus on the opinion of medical doctors that will consume these explanations. This paper presents a study asking medical professionals about their opinion of current state-of-the-art explainable artificial intelligence methods when applied to a gastrointestinal disease detection use case. We compare two different categories of explanation methods, intrinsic and extrinsic, and gauge their opinion of the current value of these explanations. The results indicate that intrinsic explanations are preferred and that physicians see value in the explanations. Based on the feedback collected in our study, future explanations of medical deep neural networks can be tailored to the needs and expectations of doctors. Hopefully, this will contribute to solving the issue of black box medical systems and lead to successful implementation of this powerful technology in the clinic.

With large chess-playing neural network models like AlphaZero contesting the state of the art within the world of computerised chess, two challenges present themselves: The question of how to explain the domain knowledge internalised by such models, and the problem that such models are not made openly available. This work presents the re-implementation of the concept detection methodology applied to AlphaZero in McGrath et al. (2022), by using large, open-source chess models with comparable performance. We obtain results similar to those achieved on AlphaZero, while relying solely on open-source resources. We also present a novel explainable AI (XAI) method, which is guaranteed to highlight exhaustively and exclusively the information used by the explained model. This method generates visual explanations tailored to domains characterised by discrete input spaces, as is the case for chess. Our presented method has the desirable property of controlling the information flow between any input vector and the given model, which in turn provides strict guarantees regarding what information is used by the trained model during inference. We demonstrate the viability of our method by applying it to standard 8x8 chess, using large open-source chess models.

The joint implementation of Federated learning (FL) and Explainable artificial intelligence (XAI) will allow training models from distributed data and explaining their inner workings while preserving important aspects of privacy. Towards establishing the benefits and tensions associated with their interplay, this scoping review maps those publications that jointly deal with FL and XAI, focusing on publications where an interplay between FL and model interpretability or post-hoc explanations was found. In total, 37 studies met our criteria, with more papers focusing on explanation methods (mainly feature relevance) than on interpretability (mainly algorithmic transparency). Most works used simulated horizontal FL setups involving 10 or fewer data centers. Only one study explicitly and quantitatively analyzed the influence of FL on model explanations, revealing a significant research gap. Aggregation of interpretability metrics across FL nodes created generalized global insights at the expense of node-specific patterns being diluted. 8 papers addressed the benefits of incorporating explanation methods as a component of the FL algorithm. Studies using established FL libraries or following reporting guidelines are a minority. More quantitative research and structured, transparent practices are needed to fully understand their mutual impact and under which conditions it happens.

This paper introduces AutoGCN, a generic Neural Architecture Search (NAS) algorithm for Human Activity Recognition (HAR) using Graph Convolution Networks (GCNs). HAR has enjoyed increased attention due to advances in deep learning, increased data availability, and enhanced computational capabilities. Concurrently, GCNs have shown promising abilities in modeling relationships between body key points in a skeletal graph. Typically, domain experts develop dataset-specific GCN-based methods, which limits their applicability beyond the specific context. AutoGCN seeks to address this limitation by simultaneously searching for the ideal hyperparameters and architecture combination within a versatile search space using a reinforcement controller while balancing optimal exploration and exploitation behavior with a knowledge reservoir during the search process. We conduct extensive experiments on two large datasets focused on skeleton-based action recognition to assess the proposed algorithm's performance. Our experimental results demonstrate the effectiveness of AutoGCN in constructing optimal GCN architectures for HAR, outperforming conventional NAS and GCN methods, as well as random search. These findings highlight the significance of a diverse search space and an expressive input representation to achieve good model performance and generalizability.

The advancement of deep learning in human activity recognition (HAR) using 3D skeleton data is critical for applications in healthcare, security, sports, and human–computer interaction. This paper tackles a well-known gap in the field, which is the lack of testing in the applicability and reliability of XAI evaluation metrics in the skeleton-based HAR domain. We have tested established XAI metrics, namely faithfulness and stability on Class Activation Mapping (CAM) and Gradient-weighted Class Activation Mapping (Grad-CAM) to address this problem. This study introduces a perturbation method that produces variations within the error tolerance of motion sensor tracking, ensuring the resultant skeletal data points remain within the plausible output range of human movement as captured by the tracking device. We used the NTU RGB+D 60 dataset and the EfficientGCN architecture for HAR model training and testing. The evaluation involved systematically perturbing the 3D skeleton data by applying controlled displacements at different magnitudes to assess the impact on XAI metric performance across multiple action classes. Our findings reveal that faithfulness may not consistently serve as a reliable metric across all classes for the EfficientGCN model, indicating its limited applicability in certain contexts. In contrast, stability proves to be a more robust metric, showing dependability across different perturbation magnitudes. Additionally, CAM and Grad-CAM yielded almost identical explanations, leading to closely similar metric outcomes. This suggests a need for the exploration of additional metrics and the application of more diverse XAI methods to broaden the understanding and effectiveness of XAI in skeleton-based HAR.