xAI 2024

Artificial intelligence has seen a significant shift in focus towards designing and developing intelligent systems that are interpretable and explainable. This is due to the complexity of the models, built from data, and the legal requirements imposed by various national and international parliaments. This has echoed both in the research literature and the press, attracting scholars worldwide and a lay audience. An emerging field with AI is eXplainable Artificial Intelligence (xAI), devoted to producing intelligent systems that allow humans to understand their inferences, assessments, predictions, recommendations and decisions. Initially devoted to designing post-hoc methods for explainability, eXplainable Artificial Intelligence (xAI) is rapidly expanding its boundaries to neuro-symbolic methods for producing self-interpretable models. Research has also shifted the focus on the structure of explanations and human-centred Artificial Intelligence since the ultimate users of interactive technologies are humans.

The World Conference on Explainable Artificial Intelligence is an annual event that aims to bring together researchers, academics, and professionals, promoting the sharing and discussion of knowledge, new perspectives, experiences, and innovations in the field of Explainable Artificial Intelligence (xAI). This event is multidisciplinary and interdisciplinary, bringing together academics and scholars of different disciplines, including Computer Science, Psychology, Philosophy, Law and Social Science, to mention a few, and industry practitioners interested in the practical, social and ethical aspects of the explanation of the models emerging from the discipline of Artificial intelligence (AI).

• Technical methods for XAI
  Action Influence Graphs  •  Agent-based explainable systems  •  Ante-hoc approaches for interpretability  •  Argumentative-based approaches for xAI  •  Argumentation theory for xAI  •  Attention mechanisms for xAI  •  Automata for explaining RNN models  •  Auto-encoders & latent spaces explainability  •  Bayesian modelling for interpretability  •  Black-boxes vs white-boxes  •  Case-based explanations for AI systems  •  Causal inference & explanations  •  Constraints-based explanations  •  Decomposition of NNET-models for XAI  •  Deep learning & XAI methods  •  Defeasible reasoning for explainability  •  Evaluation approaches for XAI-based systems  •  Explainable methods for edge computing  •  Expert systems for explainability  •  Sample-centric and dataset-centric explanations  •  Explainability of signal processing methods  •  Finite state machines for explainability  •  Fuzzy systems & logic for explainability  •  Graph neural networks for explainability  •  Hybrid & transparent black box modelling  •  Interpreting & explaining CNN Networks  •  Interpretable representational learning  •  Explainability & the Semantic Web  •  Model-specific vs model-agnostic methods  •  Neuro-symbolic reasoning for XAI  •  Natural language processing for explanations  •  Ontologies & taxonomies for supporting XAI  •  Pruning methods with XAI  •  Post-hoc methods for explainability  •  Reinforcement learning for enhancing XAI  •  Reasoning under uncertainty for explanations  •  Rule-based XAI systems  •  Robotics & explainability  •  Sample-centric & Dataset-centric explanations  •  Self-explainable methods for XAI  •  Sentence embeddings to xAI semantic features  •  Transparent & explainable learning methods  •  User interfaces for explainability  •  Visual methods for representational learning  •  XAI Benchmarking  •  XAI methods for neuroimaging & neural signals  •  XAI & reservoir computing
• Ethical Considerations for XAI
  Accountability & responsibility in XAI  •  Addressing user-centric requirements for XAI  •  Trade-off model accuracy & interpretability  •  Explainable Bias & fairness of XAI systems  •  Explainability for discovering, improving, controlling & justifying  •  Moral Principles & dilemma for XAI  •  Explainability & data fusion  •  Explainability/responsibility in policy guidelines  •  Explainability pitfalls & dark patterns in XAI  •  Historical foundations of XAI  •  Moral principles & dilemma for XAI  •  Multimodal XAI approaches  •  Philosophical consideration of synthetic explanations  •  Prevention/detection of deceptive AI explanations  •  Social implications of synthetic explanations  •  Theoretical foundations of XAI  •  Trust & explainable AI  •  The logic of scientific explanation for/in AI  •  Expected epistemic & moral goods for XAI  •  XAI for fairness checking  •  XAI for time series-based approaches
• Psychological Notions & concepts for XAI
  Algorithmic transparency & actionability  •  Cognitive approaches for explanations  •  Cognitive relief in explanations  •  Contrastive nature of explanations  •  Comprehensibility vs interpretability  •  Counterfactual explanations  •  Designing new explanation styles  •  Explanations for correctability  •  Faithfulness & intelligibility of explanations  •  Interpretability vs traceability  •  explanations Interestingness & informativeness  •  Irrelevance of probabilities to explanations  •  Iterative dialogue explanations  •  Local vs. global interpretability & explainability  •  Local vs global interpretability & explainability  •  Methods for assessing explanations quality  •  Non-technical explanations in AI systems  •  Notions and metrics of/for explainability  •  Persuasiveness & robustness of explanations  •  Psychometrics of human explanations  •  Qualitative approaches for explainability  •  Questionnaires & surveys for explainability  •  Scrutability & diagnosis of XAI methods  •  Soundness & stability of XAI methods
• Social examinations of XAI
  Adaptive explainable systems  •  Backwards & forward-looking responsibility forms to XAI  •  Data provenance & explainability  •  Explainability for reputation  •  Epistemic and non-epistemic values for XAI  •  Human-centric explainable AI  •  Person-specific XAI systems  •  Presentation & personalization of AI explanations for target groups  •  Social nature of explanations 
• Legal & administrative considerations of/for XAI
  Black-box model auditing & explanation  •  Explainability in regulatory compliance  •  Human rights for explanations in AI systems  •  Policy-based systems of explanations  •  The potential harm of explainability in AI  •  Trustworthiness of XAI for clinicians/patients  •  XAI methods for model governance  •  XAI in policy development  •  XAI for situational awareness/compliance behavior
• Safety & security approaches for XAI
  Adversarial attacks explanations  •  Explanations for risk assessment  •  Explainability of federated learning  •  Explainable IoT malware detection  •  Privacy & agency of explanations  •  XAI for Privacy-Preserving Systems  •  XAI techniques of stealing attack & defence  •  XAI for human-AI cooperation  •  XAI & models output confidence estimation
• Applications of XAI-based systems
  Application of XAI in cognitive computing  •  Dialogue systems for enhancing explainability  •    •  Explainable methods for medical diagnosis  •  Business & Marketing  •  XAI systems for healthcare  •  Explainable methods for HCI  •  Explainability in decision-support systems  •  Explainable recommender systems  •  Explainable methods for finance & automatic trading systems  •  Explainability in agricultural AI-based methods  •  Explainability in transportation systems  •  Explainability for unmanned aerial vehicles  •  Explainability in brain-computer interfaces  •  Interactive applications for XAI  •  Manufacturing chains & application of XAI  •  Models of explanations in criminology, cybersecurity & defence  •  XAI approaches in Industry 4.0  •  XAI systems for health-care  •  XAI technologies for autonomous driving  •  XAI methods for bioinformatics  •  XAI methods for linguistics/machine translation  •  XAI methods for neuroscience  •  XAI models & applications for IoT  •  XAI methods for XAI for terrestrial, atmospheric, & ocean remote sensing  •  XAI in sustainable finance & climate finance  •  XAI in bio-signals analysis