Search Results (35)

The EU Artificial Intelligence Act (Regulation (EU) 2024/1689) designates AI systems used in life and health insurance underwriting as high-risk systems, imposing rigorous requirements for bias testing, technical documentation, and post-deployment monitoring. Leveraging 12.4 million quote–bind–claim observations from four pan-European insurers (2019 Q1–2024 Q4), we evaluate how compliance affects premium schedules, loss ratios, and solvency positions. We estimate gradient-boosted decision tree (Extreme Gradient Boosting (XGBoost)) models alongside benchmark GLMs for mortality, morbidity, and lapse risk, using Shapley Additive Explanations (SHAP) values for explainability. Protected attributes (gender, ethnicity proxy, disability, and postcode deprivation) are excluded from training but retained for audit. We measure bias via statistical parity difference, disparate impact ratio, and equalized odds gap against the 10 percent tolerance in regulatory guidance, and then apply counterfactual mitigation strategies—re-weighing, reject option classification, and adversarial debiasing. We simulate impacts on expected loss ratios, the Solvency II Standard Formula Solvency Capital Requirement (SCR), and internal model economic capital. To translate fairness breaches into compliance risk, we compute expected penalties under the Act’s two-tier fine structure and supervisory detection probabilities inferred from GDPR enforcement. Under stress scenarios—full retraining, feature excision, and proxy disclosure—preliminary results show that bottom-income quintile premiums exceed fair benchmarks by 5.8 percent (life) and 7.2 percent (health). Mitigation closes 65–82 percent of these gaps but raises capital requirements by up to 4.1 percent of own funds; expected fines exceed rectification costs once detection probability surpasses 9 percent. We conclude that proactive adversarial debiasing offers insurers a capital-efficient compliance pathway and outline implications for enterprise risk management and future monitoring. Full article

Large Language Models (LLMs) excel in diverse text generation tasks but still face limited controllability, opaque decision processes, and frequent hallucinations. This paper presents a structural causal intervention framework that models input–hidden–output dependencies through a structural causal model and performs targeted interventions on hidden representations. By combining counterfactual sample construction with contrastive training, our method enables precise control of style, sentiment, and factual consistency while providing explicit causal explanations for output changes. Experiments on three representative tasks demonstrate consistent and substantial improvements: style transfer accuracy reaches 92.3% (+7–14 percentage points over strong baselines), sentiment-controlled generation achieves 90.1% accuracy (+1.3–10.9 points), and multi-attribute conflict rates drop to 3.7% (a 40–60% relative reduction). Our method also improves causal attribution scores to 0.83–0.85 and human agreement rates to 87–88%, while reducing training and inference latency by 25–30% through sparse masking that modifies ≤10% of hidden units per attribute. These results confirm that integrating structural causal intervention with counterfactual training advances controllability, interpretability, and efficiency in LLM-based generation, offering a robust foundation for deployment in reliability-critical and resource-constrained applications. Full article

For the decision-making challenge of ensuring vehicle lane-changing safety, this study proposes a context-dependent causal inference-based model for safe lane changes. Emphasizing multi-vehicle interactions within dynamic traffic scenarios, we construct a three-layer decision-making framework that relies on real-time data collection of speed, acceleration, and spacing information from both the target vehicle and adjacent-lane vehicles. The framework consists of (1) a context-aware layer that extracts standardized dynamic features; (2) an attention mechanism layer that dynamically assigns weights to critical risk factors; and (3) a counterfactual causal reasoning layer where lane-changing risks are quantified through virtual interventions, with multi-objective safety strategies optimized via particle swarm algorithms. The simulation results indicate significant enhancements in high-density traffic conditions. When compared to traditional safety distance models and built-in models from simulation software (SUMO v1.18.0), the proposed model achieves reductions in average conflict counts by 63.0% (from 12.7 to 4.7 instances) and by 37.3% (from 7.5 to 4.7 instances), respectively. Additionally, lane-changing durations are reduced by 10.9% (from 5.5 to 4.9 s) and by 31.9% (from 7.2 to 4.9 s), while fluctuations in risk values decrease by 53.3% (from 0.75 to 0.35) and by 36.4% (from 0.55 to 0.35), respectively. The experimental validation confirms that the integration of dynamic safety distance computation with causal reasoning significantly enhances decision-making robustness in complex scenarios through coordinated risk quantification and multi-objective optimization Full article

We propose a novel deep learning framework for counterfactual inference on the COMPAS dataset, utilizing a multi-task CNN-LSTM architecture. The model jointly predicts multiple outcome types: (i) count outcomes with zero inflation, modeled using zero-inflated Poisson (ZIP), zero-inflated negative binomial (ZINB), and negative binomial (NB) distributions; (ii) time-to-event outcomes, modeled via the Cox proportional hazards model. To effectively leverage the structure in high-dimensional tabular data, we integrate functional data analysis (FDA) techniques by transforming covariates into smooth functional representations using B-spline basis expansions. Specifically, we construct a pseudo-temporal index over predictor variables and fit basis expansions to each subject’s feature vector, yielding a low-dimensional set of coefficients that preserve smooth variation while reducing noise. This functional representation enables the CNN-LSTM model to capture both local and global temporal patterns in the data, including treatment-covariate interactions. Our approach estimates both population-average and individual-level treatment effects (ATE and CATE) for each outcome and evaluates predictive performance using metrics such as Poisson deviance, root mean squared error (RMSE), and the concordance index (C-index). Statistical inference on treatment effects is supported via bootstrap-based confidence intervals and hypothesis testing. Overall, this comprehensive framework facilitates flexible modeling of heterogeneous treatment effects in structured, high-dimensional data, advancing causal inference methodologies in criminal justice and related domains. Full article

Objectives: We introduce the conceptual framework for the Adversarial Claim Robustness Diagnostics (ACRD) protocol, a novel tool for assessing how factual claims withstand ideological distortion. Methods: Based on semantics, adversarial collaboration, and the devil’s advocate approach, we develop a three-phase evaluation process combining baseline evaluations, adversarial speaker reframing, and dynamic AI calibration along with quantified robustness scoring. We introduce the Claim Robustness Index that constitutes our final validity scoring measure. Results: We model the evaluation of claims by ideologically opposed groups as a strategic game with a Bayesian-Nash equilibrium to infer the normative behavior of evaluators after the reframing phase. The ACRD addresses shortcomings in traditional fact-checking approaches and employs large language models to simulate counterfactual attributions while mitigating potential biases. Conclusions: The framework’s ability to identify boundary conditions of persuasive validity across polarized groups can be tested across important societal and political debates ranging from climate change issues to trade policy discourses. Full article

Video-based person re-identification (re-identification) remains underused in real-world deployments, despite impressive benchmark performance. Most existing models rely on superficial correlations—such as clothing, background, or lighting—that fail to generalize across domains, viewpoints, and temporal variations. This study examines the emerging role of causal reasoning as a principled alternative to traditional correlation-based approaches in video-based re-identification. We provide a structured and critical analysis of methods that leverage structural causal models (SCMs), interventions, and counterfactual reasoning to isolate identity-specific features from confounding factors. This study is organized around a novel taxonomy of causal re-identification methods, spanning generative disentanglement, domain-invariant modeling, and causal transformers. We review current evaluation metrics and introduce causal-specific robustness measures. In addition, we assess the practical challenges—scalability, fairness, interpretability, and privacy—that must be addressed for real-world adoption. Finally, we identify open problems and outline future research directions that integrate causal modeling with efficient architectures and self-supervised learning. This study aims to establish a coherent foundation for causal video-based person re-identification and catalyze the next phase of research in this rapidly evolving domain. Full article

Complex question decomposition is an important research topic in the field of natural language processing (NLP). It refers to the decomposition of a compound question containing multiple ontologies and classes into a simple question containing only a single attribute or entity. Most previous studies focus on how to generate simple questions using a single attribute or entity but pay little attention to the generation order of simple questions, which may lead to an inaccurate decomposition or longer execution time. In this study, we propose a new method based on causal reinforcement learning, which combines the advantages of the current optimal performance reinforcement learning method and the causal inference method. Compared with previous methods, causal reinforcement learning can find the generation order of sub-questions more accurately, so as to better decompose complex questions. In particular, the prior knowledge is extracted using the counterfactual method in causal reasoning and is integrated into the policy network of the reinforcement learning model, and the reward rules of reinforcement learning are designed from the perspective of symmetry (positive reward and negative punishment), thus the intelligent body is guided to choose the sub-question with a greater benefit and less risk of decomposing. We compare the proposed method with the baseline method on three datasets. The experimental results show that the performance of our method is improved by 5–10% compared with the baseline method on Hits@n (n = 1, 3, 10), which proves the effectiveness of our proposed method. Full article

The rapid expansion of urban rail networks has resulted in increasingly complex passenger flow patterns, presenting significant challenges for operational management, especially during incidents and emergencies. Disruptions such as power equipment failures, trackside faults, and train malfunctions can severely impact transit efficiency and reliability, leading to congestion and cascading network effects. Existing models for predicting passenger origin–destination (OD) matrices struggle to provide accurate and timely predictions under these disrupted conditions. This study proposes a deep counterfactual inference model that improves both the prediction accuracy and interpretability of OD matrices during incidents. The model uses a dual-channel framework based on multi-task learning, where the factual channel predicts OD matrices under normal conditions and the counterfactual channel estimates OD matrices during incidents, enabling the quantification of the spatiotemporal impacts of disruptions. Our approach which incorporates KL divergence-based propensity matching enhances prediction accuracy by 4.761% to 12.982% compared to baseline models, while also providing interpretable insights into disruption mechanisms. The model reveals that incident types vary in delay magnitude, with power equipment incidents causing the largest delays, and shows that incidents have time-lag effects on OD flows, with immediate impacts on origin stations and progressively delayed effects on destination and neighboring stations. This research offers practical tools for urban rail transit operators to estimate incident-affected passenger volumes and implement more efficient emergency response strategies, advancing emergency response capabilities in smart transit systems. Full article

Multimodal emotion recognition (MER) aims to enhance the understanding of human emotions by integrating visual, auditory, and textual modalities. However, previous MER approaches often depend on a dominant modality rather than considering all modalities, leading to poor generalization. To address this, we propose Causal Inference in Multimodal Emotion Recognition (CausalMER), which leverages counterfactual reasoning and causal graphs to capture relationships between modalities and reduce direct modality effects contributing to bias. This allows CausalMER to make unbiased predictions while being easily applied to existing MER methods in a model-agnostic manner, without requiring any architectural modifications. We evaluate CausalMER on the IEMOCAP and CMU-MOSEI datasets, widely used benchmarks in MER, and compare it with existing methods. On the IEMOCAP dataset with the MulT backbone, CausalMER achieves an average accuracy of 83.4%. On the CMU-MOSEI dataset, the average accuracies with MulT, PMR, and DMD backbones are 50.1%, 48.8%, and 48.8%, respectively. Experimental results demonstrate that CausalMER is robust in missing modality scenarios, as shown by its low standard deviation in performance drop gaps. Additionally, we evaluate modality contributions and show that CausalMER achieves balanced contributions from each modality, effectively mitigating direct biases from individual modalities. Full article

Given the rapid advancement of artificial intelligence, understanding the foundations of intelligent behaviour is increasingly important. Active inference, regarded as a general theory of behaviour, offers a principled approach to probing the basis of sophistication in planning and decision-making. This paper examines two decision-making schemes in active inference based on “planning” and “learning from experience”. Furthermore, we also introduce a mixed model that navigates the data complexity trade-off between these strategies, leveraging the strengths of both to facilitate balanced decision-making. We evaluate our proposed model in a challenging grid-world scenario that requires adaptability from the agent. Additionally, our model provides the opportunity to analyse the evolution of various parameters, offering valuable insights and contributing to an explainable framework for intelligent decision-making. Full article

As the world’s largest emerging market country, not only has China faced the contradiction between its huge population size and per capita energy scarcity for a long time, but the rigid constraints brought by energy poverty have also plagued the lives and production of Chinese residents. Based on panel data from 30 provinces (except Tibet) in mainland China from 2009 to 2021, this study employs double machine learning and spatial difference-in-difference for causal inference to explore the impact of a medium- to long-term regional innovation pilot policy in China—the new policy for innovative transformation in regional industrial chains—on energy poverty alleviation. This study also introduces China’s conversion of new and old kinetic energy into this quasi-natural experiment. This study presents the following findings: (1) The new policy for innovative transformation in regional industrial chains and the concept of the conversion of new and old kinetic energy can both significantly promote energy poverty alleviation. (2) The mechanism pathway of “the new policy for innovative transformation in regional industrial chains → the conversion of new and old kinetic energy → the energy poverty alleviation in heating/household electricity/transportation segments” has proved to be an effective practice in China. (3) Based on the spatial double difference model, the spatial direct effect of the new regional industrial chain innovation and change policy on energy poverty alleviation is significantly positive, while the spatial direct effect and spatial spillover effect of the new and old kinetic energy transformation on energy poverty alleviation are both significantly positive. (4) Based on the counterfactual framework analysis, in addition to the causal mediating mechanism of the demand-side conversion of new and old kinetic energy being impeded, both the supply-side and the structural-side conversion of new and old kinetic energy are able to play a significant positive causal mediating role in both the treatment and control groups. Full article

Most mainstream statistical models will achieve poor performance in Out-Of-Distribution (OOD) generalization. This is because these models tend to learn the spurious correlation between data and will collapse when the domain shift exists. If we want artificial intelligence (AI) to make great strides in real life, the current focus needs to be shifted to the OOD problem of deep learning models to explore the generalization ability under unknown environments. Domain generalization (DG) focusing on OOD generalization is proposed, which is able to transfer the knowledge extracted from multiple source domains to the unseen target domain. We are inspired by intuitive thinking about human intelligence relying on causality. Unlike relying on plain probability correlations, we apply a novel causal perspective to DG, which can improve the OOD generalization ability of the trained model by mining the invariant causal mechanism. Firstly, we construct the inclusive causal graph for most DG tasks through stepwise causal analysis based on the data generation process in the natural environment and introduce the reasonable Structural Causal Model (SCM). Secondly, based on counterfactual inference, causal semantic representation learning with domain intervention (CSRDN) is proposed to train a robust model. In this regard, we generate counterfactual representations for different domain interventions, which can help the model learn causal semantics and develop generalization capacity. At the same time, we seek the Pareto optimal solution in the optimization process based on the loss function to obtain a more advanced training model. Extensive experimental results of Rotated MNIST and PACS as well as VLCS datasets verify the effectiveness of the proposed CSRDN. The proposed method can integrate causal inference into domain generalization by enhancing interpretability and applicability and brings a boost to challenging OOD generalization problems. Full article

Well-defined and stable property rights play a pivotal role in shaping human economic behavior by averting the tragedy of the commons. This study employs micro-survey data from Heilongjiang Province, China, to empirically investigate the impact and mechanisms of land approval on the adoption of straw-returning technology by farmers. Utilizing the Probit model and mediation and moderation effect testing methods, the findings reveal the following: (1) Land approval significantly promotes the adoption of straw-returning techniques by farmers, with a marginal effect of 0.288. This view is further validated through counterfactual inference constructed using the propensity score matching method. (2) Endowment effects mediate the relationship between land approval and farmers’ adoption of straw-returning technology. (3) Digital skills and farming scale negatively moderate the policy’s impact on farmers’ adoption of straw-returning technology. (4) In terms of control variables, the age of farmers and the dispersion of cultivated land have a significant negative impact on the adoption of straw-returning technology by farmers, while training related to agricultural straw-returning skills and government technology promotion significantly positively affects the use of straw-returning technology by farmers. Therefore, the clarity of land property rights helps to harness the policy effects of land approval and provides a research approach for countries with communal land ownership to implement actions for soil quality conservation. Full article

The prediction of building energy consumption is beneficial to utility companies, users, and facility managers to reduce energy waste. However, due to various drawbacks of prediction algorithms, such as, non-transparent output, ad hoc explanation by post hoc tools, low accuracy, and the inability to deal with data uncertainties, such prediction has limited applicability in this domain. As a result, domain knowledge-based explainability with high accuracy is critical for making energy predictions trustworthy. Motivated by this, we propose an advanced explainable Belief Rule-Based Expert System (eBRBES) with domain knowledge-based explanations for the accurate prediction of energy consumption. We optimize BRBES’s parameters and structure to improve prediction accuracy while dealing with data uncertainties using its inference engine. To predict energy consumption, we take into account floor area, daylight, indoor occupancy, and building heating method. We also describe how a counterfactual output on energy consumption could have been achieved. Furthermore, we propose a novel Belief Rule-Based adaptive Balance Determination (BRBaBD) algorithm for determining the optimal balance between explainability and accuracy. To validate the proposed eBRBES framework, a case study based on Skellefteå, Sweden, is used. BRBaBD results show that our proposed eBRBES framework outperforms state-of-the-art machine learning algorithms in terms of optimal balance between explainability and accuracy by 85.08%. Full article

In this study, we (a) formulated a general hypothesis about how wetland (functional and structural) traits influence avian diversity, (b) turned this hypothesis into a non-parametric Bayesian network, (c) disentangled the direct and indirect effects of the variables influencing waterbird species, and (d) simulated the changes expected to the levels of avian diversity as a result of numerous counterfactual and management scenarios. We applied our framework to the Sicilian wetlands as a whole; then, we downscaled simulations locally to a wetland of particular interest (Pantano Bruno). We found that (1) waterbird species are highly sensitive to wetland traits; (2) wetland traits have both direct and indirect effects upon alpha avian diversity; (3) the direct and indirect effects of wetland traits can be contrasting; (4) water level fluctuations (benefit), diversions (cost), and salinity (cost) are key factors for waterbird conservation; (5) these wetlands have the potential for hosting a level of alpha avian diversity that is double the baseline (from 19 to 38 species); (6) these wetlands are prone to ecological collapse if all traits deteriorate (from 19 to 6 species per wetland); and (7) the ecological information gained at the regional scale can be properly downscaled to the local scale to make inferences on single wetlands. Full article