Research Papers

Explainable Artificial Intelligence (XAI) aims to make learning machines less opaque, and offers researchers and practitioners various tools to reveal the decision-making strategies of neural networks. In this work, we investigate how XAI methods can be used for exploring and visualizing the diversity of feature representations learned by Bayesian Neural Networks (BNNs). Our goal is to provide a global understanding of BNNs by making their decision-making strategies a) visible and tangible through feature visualizations and b) quantitatively measurable with a distance measure learned by contrastive learning. Our work provides new insights into the \emph{posterior} distribution in terms of human-understandable feature information with regard to the underlying decision making strategies. The main findings of our work are the following: 1) global XAI methods can be applied to explain the diversity of decision-making strategies of BNN instances, 2) Monte Carlo dropout with commonly used Dropout rates exhibit increased diversity in feature representations compared to the multimodal posterior approximation of MultiSWAG, 3) the diversity of learned feature representations highly correlates with the uncertainty estimate for the output and 4) the inter-mode diversity of the multimodal posterior decreases as the network width increases, while the intra mode diversity increases. These findings are consistent with the recent Deep Neural Networks theory, providing additional intuitions about what the theory implies in terms of humanly understandable concepts.

The issue of fairness in machine learning stems from the fact that historical data often displays biases against specific groups of people which have been underprivileged in the recent past, or still are. In this context, one of the possible approaches is to employ fair representation learning algorithms which are able to remove biases from data, making groups statistically indistinguishable. In this paper, we instead develop a fair representation learning algorithm which is able to map individuals belonging to different groups in a single group. This is made possible by training a pair of Normalizing Flow models and constraining them to not remove information about the ground truth by training a ranking or classification model on top of them. The overall, ``chained'' model is invertible and has a tractable Jacobian, which allows to relate together the probability densities for different groups and ``translate'' individuals from one group to another. We show experimentally that our methodology is competitive with other fair representation learning algorithms. Furthermore, our algorithm achieves stronger invariance w.r.t. the sensitive attribute.

A key issue in critical contexts such as medical diagnosis is the interpretability of the deep learning models adopted in decision-making systems. Research in eXplainable Artificial Intelligence (XAI) is trying to solve this issue. However, often XAI approaches are only tested on generalist classifier and do not represent realistic problems such as those of medical diagnosis. In this paper, we analyze a case study on skin lesion images where we customize an existing XAI approach for explaining a deep learning model able to recognize different types of skin lesions. The explanation is formed by synthetic exemplar and counter-exemplar images of skin lesion and offers the practitioner a way to highlight the crucial traits responsible for the classification decision. A survey conducted with domain experts, beginners and unskilled people proof that the usage of explanations increases the trust and confidence in the automatic decision system. Also, an analysis of the latent space adopted by the explainer unveils that some of the most frequent skin lesion classes are distinctly separated. This phenomenon could derive from the intrinsic characteristics of each class and, hopefully, can provide support in the resolution of the most frequent misclassifications by human experts.

Robots of the future are going to exhibit increasingly human-like and super-human intelligence in a myriad of different tasks. They are also likely going to fail and be incompliant with human preferences in increasingly subtle ways. Towards the goal of achieving autonomous robots, the robot learning community has made rapid strides in applying machine learning techniques to train robots through data and interaction. This makes the study of how best to audit these algorithms for checking their compatibility with humans, pertinent and urgent. In this paper, we draw inspiration from the AI Safety and Alignment communities and make the case that we need to urgently consider ways in which we can best audit our robot learning algorithms to check for failure modes, and ensure that when operating autonomously, they are indeed behaving in ways that the human algorithm designers intend them to. We believe that this is a challenging problem that will require efforts from the entire robot learning community, and do not attempt to provide a concrete framework for auditing. Instead, we outline high-level guidance and a possible approach towards formulating this framework which we hope will serve as a useful starting point for thinking about auditing in the context of robot learning.

The results showed that XAI could indeed be used as a proof of concept to begin discussions on the implementation of assistive AI systems within the clinical community.

The increasingly widespread application of AI models motivates increased demand for explanations from a variety of stakeholders. However, this demand is ambiguous because there are many types of 'explanation' with different evaluative criteria. In the spirit of pluralism, I chart a taxonomy of types of explanation and the associated XAI methods that can address them. When we look to expose the inner mechanisms of AI models, we develop Diagnostic-explanations. When we seek to render model output understandable, we produce Explication-explanations. When we wish to form stable generalizations of our models, we produce Expectation-explanations. Finally, when we want to justify the usage of a model, we produce Role-explanations that situate models within their social context. The motivation for such a pluralistic view stems from a consideration of causes as manipulable relationships and the different types of explanations as identifying the relevant points in AI systems we can intervene upon to affect our desired changes. This paper reduces the ambiguity in use of the word 'explanation' in the field of XAI, allowing practitioners and stakeholders a useful template for avoiding equivocation and evaluating XAI methods and putative explanations.

The black-box nature of machine learning models limits their use in case-critical applications, raising faithful and ethical concerns that lead to trust crises. One possible way to mitigate this issue is to understand how a (mispredicted) decision is carved out from the decision boundary. This paper presents a human-in-the-loop approach to explain machine learning models using verbatim neighborhood manifestation. Contrary to most of the current eXplainable Artificial Intelligence (XAI) systems, which provide hit-or-miss approximate explanations, our approach generates the local decision boundary of the given instance and enables human intelligence to conclude the model behavior. Our method can be divided into three stages: 1) a neighborhood generation stage, which generates instances based on the given sample; 2) a classification stage, which yields classifications on the generated instances to carve out the local decision boundary and delineate the model behavior; and 3) a human-in-the-loop stage, which involves human to refine and explore the neighborhood of interest. In the generation stage, a generative model is used to generate the plausible synthetic neighbors around the given instance. After the classification stage, the classified neighbor instances provide a multifaceted understanding of the model behavior. Three intervention points are provided in the human-in-the-loop stage, enabling humans to leverage their own intelligence to interpret the model behavior. Several experiments on two datasets are conducted, and the experimental results demonstrate the potential of our proposed approach for boosting human understanding of the complex machine learning model.

Algorithmic Insurance

Dimitris Bertsimas, Agni Orfanoudaki

When AI systems make errors in high-stakes domains like medical diagnosis or autonomous vehicles, a single algorithmic flaw across varying operational contexts can generate highly heterogeneous losses that challenge traditional insurance assumptions. Algorithmic insurance constitutes a novel form of financial coverage for AI-induced damages, representing an emerging market that addresses algorithm-driven liability. However, insurers currently struggle to price these risks, while AI developers lack rigorous frameworks connecting system design with financial liability exposure. We analyze the connection between operational choices of binary classification performance to tail risk exposure. Using conditional value-at-risk (CVaR) to capture extreme losses, we prove that established approaches like maximizing accuracy can significantly increase worst-case losses compared to tail risk optimization, with penalties growing quadratically as thresholds deviate from optimal. We then propose a liability insurance contract structure that mandates risk-aware classification thresholds and characterize the conditions under which it creates value for AI providers. Our analysis extends to degrading model performance and human oversight scenarios. We validate our findings through a mammography case study, demonstrating that CVaR-optimal thresholds reduce tail risk up to 13-fold compared to accuracy maximization. This risk reduction enables insurance contracts to create 14-16% gains for well-calibrated firms, while poorly calibrated firms benefit up to 65% through risk transfer, mandatory recalibration, and regulatory capital relief. Unlike traditional insurance that merely transfers risk, algorithmic insurance can function as both a financial instrument and an operational governance mechanism, simultaneously enabling efficient risk transfer while improving AI safety.

Designer-User Communication for XAI: An epistemological approach to discuss XAI design

Juliana Jansen Ferreira, Mateus Monteiro

Artificial Intelligence is becoming part of any technology we use nowadays. If the AI informs people's decisions, the explanation about AI's outcomes, results, and behavior becomes a necessary capability. However, the discussion of XAI features with various stakeholders is not a trivial task. Most of the available frameworks and methods for XAI focus on data scientists and ML developers as users. Our research is about XAI for end-users of AI systems. We argue that we need to discuss XAI early in the AI-system design process and with all stakeholders. In this work, we aimed at investigating how to operationalize the discussion about XAI scenarios and opportunities among designers and developers of AI and its end-users. We took the Signifying Message as our conceptual tool to structure and discuss XAI scenarios. We experiment with its use for the discussion of a healthcare AI-System.

Growing concerns regarding the operational usage of AI models in the real-world has caused a surge of interest in explaining AI models' decisions to humans. Reinforcement Learning is not an exception in this regard. In this work, we propose a method for offering local explanations on risk in reinforcement learning. Our method only requires a log of previous interactions between the agent and the environment to create a state-transition model. It is designed to work on RL environments with either continuous or discrete state and action spaces. After creating the model, actions of any agent can be explained in terms of the features most influential in increasing or decreasing risk or any other desirable objective function in the locality of the agent. Through experiments, we demonstrate the effectiveness of the proposed method in providing such explanations.

Twin Systems for DeepCBR: A Menagerie of Deep Learning and Case-Based Reasoning Pairings for Explanation and Data Augmentation

Mark T Keane, Eoin M Kenny, Mohammed Temraz +2

Recently, it has been proposed that fruitful synergies may exist between Deep Learning (DL) and Case Based Reasoning (CBR); that there are insights to be gained by applying CBR ideas to problems in DL (what could be called DeepCBR). In this paper, we report on a program of research that applies CBR solutions to the problem of Explainable AI (XAI) in the DL. We describe a series of twin-systems pairings of opaque DL models with transparent CBR models that allow the latter to explain the former using factual, counterfactual and semi-factual explanation strategies. This twinning shows that functional abstractions of DL (e.g., feature weights, feature importance and decision boundaries) can be used to drive these explanatory solutions. We also raise the prospect that this research also applies to the problem of Data Augmentation in DL, underscoring the fecundity of these DeepCBR ideas.

Handling Climate Change Using Counterfactuals: Using Counterfactuals in Data Augmentation to Predict Crop Growth in an Uncertain Climate Future

Mohammed Temraz, Eoin Kenny, Elodie Ruelle +3

Climate change poses a major challenge to humanity, especially in its impact on agriculture, a challenge that a responsible AI should meet. In this paper, we examine a CBR system (PBI-CBR) designed to aid sustainable dairy farming by supporting grassland management, through accurate crop growth prediction. As climate changes, PBI-CBRs historical cases become less useful in predicting future grass growth. Hence, we extend PBI-CBR using data augmentation, to specifically handle disruptive climate events, using a counterfactual method (from XAI). Study 1 shows that historical, extreme climate-events (climate outlier cases) tend to be used by PBI-CBR to predict grass growth during climate disrupted periods. Study 2 shows that synthetic outliers, generated as counterfactuals on a outlier-boundary, improve the predictive accuracy of PBICBR, during the drought of 2018. This study also shows that an instance-based counterfactual method does better than a benchmark, constraint-guided method.