Research Papers

The applications of Artificial Intelligence (AI) surround decisions on increasingly many aspects of human lives. Society responds by imposing legal and social expectations for the accountability of such automated decision systems (ADSs). Fairness, a fundamental constituent of AI accountability, is concerned with just treatment of individuals and sensitive groups (e.g., based on sex, race). While many studies focus on fair learning and fairness testing for the classification tasks, the literature is rather limited on how to examine fairness in regression tasks. This work presents error parity as a regression fairness notion and introduces a testing methodology to assess group fairness based on a statistical hypothesis testing procedure. The error parity test checks whether prediction errors are distributed similarly across sensitive groups to determine if an ADS is fair. It is followed by a suitable permutation test to compare groups on several statistics to explore disparities and identify impacted groups. The usefulness and applicability of the proposed methodology are demonstrated via a case study on COVID-19 projections in the US at the county level, which revealed race-based differences in forecast errors. Overall, the proposed regression fairness testing methodology fills a gap in the fair machine learning literature and may serve as a part of larger accountability assessments and algorithm audits.

On the explainable properties of 1-Lipschitz Neural Networks: An Optimal Transport Perspective

Mathieu Serrurier, Franck Mamalet, Thomas Fel +2

Input gradients have a pivotal role in a variety of applications, including adversarial attack algorithms for evaluating model robustness, explainable AI techniques for generating Saliency Maps, and counterfactual explanations.However, Saliency Maps generated by traditional neural networks are often noisy and provide limited insights. In this paper, we demonstrate that, on the contrary, the Saliency Maps of 1-Lipschitz neural networks, learned with the dual loss of an optimal transportation problem, exhibit desirable XAI properties:They are highly concentrated on the essential parts of the image with low noise, significantly outperforming state-of-the-art explanation approaches across various models and metrics. We also prove that these maps align unprecedentedly well with human explanations on ImageNet.To explain the particularly beneficial properties of the Saliency Map for such models, we prove this gradient encodes both the direction of the transportation plan and the direction towards the nearest adversarial attack. Following the gradient down to the decision boundary is no longer considered an adversarial attack, but rather a counterfactual explanation that explicitly transports the input from one class to another. Thus, Learning with such a loss jointly optimizes the classification objective and the alignment of the gradient, i.e. the Saliency Map, to the transportation plan direction.These networks were previously known to be certifiably robust by design, and we demonstrate that they scale well for large problems and models, and are tailored for explainability using a fast and straightforward method.

BERTuit: Understanding Spanish language in Twitter through a native transformer

Javier Huertas-Tato, Alejandro Martin, David Camacho

The appearance of complex attention-based language models such as BERT, Roberta or GPT-3 has allowed to address highly complex tasks in a plethora of scenarios. However, when applied to specific domains, these models encounter considerable difficulties. This is the case of Social Networks such as Twitter, an ever-changing stream of information written with informal and complex language, where each message requires careful evaluation to be understood even by humans given the important role that context plays. Addressing tasks in this domain through Natural Language Processing involves severe challenges. When powerful state-of-the-art multilingual language models are applied to this scenario, language specific nuances use to get lost in translation. To face these challenges we present \textbf{BERTuit}, the larger transformer proposed so far for Spanish language, pre-trained on a massive dataset of 230M Spanish tweets using RoBERTa optimization. Our motivation is to provide a powerful resource to better understand Spanish Twitter and to be used on applications focused on this social network, with special emphasis on solutions devoted to tackle the spreading of misinformation in this platform. BERTuit is evaluated on several tasks and compared against M-BERT, XLM-RoBERTa and XLM-T, very competitive multilingual transformers. The utility of our approach is shown with applications, in this case: a zero-shot methodology to visualize groups of hoaxes and profiling authors spreading disinformation. Misinformation spreads wildly on platforms such as Twitter in languages other than English, meaning performance of transformers may suffer when transferred outside English speaking communities.

The new era of image segmentation leveraging the power of Deep Neural Nets (DNNs) comes with a price tag: to train a neural network for pixel-wise segmentation, a large amount of training samples has to be manually labeled on pixel-precision. In this work, we address this by following an indirect solution. We build upon the advances of the Explainable AI (XAI) community and extract a pixel-wise binary segmentation from the output of the Layer-wise Relevance Propagation (LRP) explaining the decision of a classification network. We show that we achieve similar results compared to an established U-Net segmentation architecture, while the generation of the training data is significantly simplified. The proposed method can be trained in a weakly supervised fashion, as the training samples must be only labeled on image-level, at the same time enabling the output of a segmentation mask. This makes it especially applicable to a wider range of real applications where tedious pixel-level labelling is often not possible.

Monumental advances in deep learning have led to unprecedented achievements across various domains. While the performance of deep neural networks is indubitable, the architectural design and interpretability of such models are nontrivial. Research has been introduced to automate the design of neural network architectures through neural architecture search (NAS). Recent progress has made these methods more pragmatic by exploiting distributed computation and novel optimization algorithms. However, there is little work in optimizing architectures for interpretability. To this end, we propose a multi-objective distributed NAS framework that optimizes for both task performance and "introspectability," a surrogate metric for aspects of interpretability. We leverage the non-dominated sorting genetic algorithm (NSGA-II) and explainable AI (XAI) techniques to reward architectures that can be better comprehended by domain experts. The framework is evaluated on several image classification datasets. We demonstrate that jointly optimizing for task error and introspectability leads to more disentangled and debuggable architectures that perform within tolerable error.

Face morphing attacks can compromise Face Recognition System (FRS) by exploiting their vulnerability. Face Morphing Attack Detection (MAD) techniques have been developed in recent past to deter such attacks and mitigate risks from morphing attacks. MAD algorithms, as any other algorithms should treat the images of subjects from different ethnic origins in an equal manner and provide non-discriminatory results. While the promising MAD algorithms are tested for robustness, there is no study comprehensively bench-marking their behaviour against various ethnicities. In this paper, we study and present a comprehensive analysis of algorithmic fairness of the existing Single image-based Morph Attack Detection (S-MAD) algorithms. We attempt to better understand the influence of ethnic bias on MAD algorithms and to this extent, we study the performance of MAD algorithms on a newly created dataset consisting of four different ethnic groups. With Extensive experiments using six different S-MAD techniques, we first present benchmark of detection performance and then measure the quantitative value of the algorithmic fairness for each of them using Fairness Discrepancy Rate (FDR). The results indicate the lack of fairness on all six different S-MAD methods when trained and tested on different ethnic groups suggesting the need for reliable MAD approaches to mitigate the algorithmic bias.

Advances in machine learning (ML) technologies have greatly improved Artificial Intelligence (AI) systems. As a result, AI systems have become ubiquitous, with their application prevalent in virtually all sectors. However, AI systems have prompted ethical concerns, especially as their usage crosses boundaries in sensitive areas such as healthcare, transportation, and security. As a result, users are calling for better AI governance practices in ethical AI systems. Therefore, AI development methods are encouraged to foster these practices. This research analyzes the ECCOLA method for developing ethical and trustworthy AI systems to determine if it enables AI governance in development processes through ethical practices. The results demonstrate that while ECCOLA fully facilitates AI governance in corporate governance practices in all its processes, some of its practices do not fully foster data governance and information governance practices. This indicates that the method can be further improved.

JEDAI: A System for Skill-Aligned Explainable Robot Planning

Naman Shah, Pulkit Verma, Trevor Angle +1

This paper presents JEDAI, an AI system designed for outreach and educational efforts aimed at non-AI experts. JEDAI features a novel synthesis of research ideas from integrated task and motion planning and explainable AI. JEDAI helps users create high-level, intuitive plans while ensuring that they will be executable by the robot. It also provides users customized explanations about errors and helps improve their understanding of AI planning as well as the limits and capabilities of the underlying robot system.

Principles of fairness and solidarity in AI ethics regularly overlap, creating obscurity in practice: acting in accordance with one can appear indistinguishable from deciding according to the rules of the other. However, there exist irregular cases where the two concepts split, and so reveal their disparate meanings and uses. This paper explores two cases in AI medical ethics, one that is irregular and the other more conventional, to fully distinguish fairness and solidarity. Then the distinction is applied to the frequently cited COMPAS versus ProPublica dispute in judicial ethics. The application provides a broader model for settling contemporary and topical debates about fairness and solidarity. It also implies a deeper and disorienting truth about AI ethics principles and their justification.

To ensure trust in AI models, it is becoming increasingly apparent that evaluation of models must be extended beyond traditional performance metrics, like accuracy, to other dimensions, such as fairness, explainability, adversarial robustness, and distribution shift. We describe an empirical study to evaluate multiple model types on various metrics along these dimensions on several datasets. Our results show that no particular model type performs well on all dimensions, and demonstrate the kinds of trade-offs involved in selecting models evaluated along multiple dimensions.

The field explainable artificial intelligence (XAI) has brought about an arsenal of methods to render Machine Learning (ML) predictions more interpretable. But how useful explanations provided by transparent ML methods are for humans remains difficult to assess. Here we investigate the quality of interpretable computer vision algorithms using techniques from psychophysics. In crowdsourced annotation tasks we study the impact of different interpretability approaches on annotation accuracy and task time. We compare these quality metrics with classical XAI, automated quality metrics. Our results demonstrate that psychophysical experiments allow for robust quality assessment of transparency in machine learning. Interestingly the quality metrics computed without humans in the loop did not provide a consistent ranking of interpretability methods nor were they representative for how useful an explanation was for humans. These findings highlight the potential of methods from classical psychophysics for modern machine learning applications. We hope that our results provide convincing arguments for evaluating interpretability in its natural habitat, human-ML interaction, if the goal is to obtain an authentic assessment of interpretability.

Algorithmic Bias can be due to bias in the training data or issues with the algorithm itself. These algorithmic issues typically relate to problems with model capacity and regularisation. This underestimation bias may arise because the model has been optimised for good generalisation accuracy without any explicit consideration of bias or fairness. In a sense, we should not be surprised that a model might be biased when it hasn't been "asked" not to be. In this paper, we consider including bias (underestimation) as an additional criterion in model training. We present a multi-objective optimisation strategy using Pareto Simulated Annealing that optimise for both balanced accuracy and underestimation. We demonstrate the effectiveness of this strategy on one synthetic and two real-world datasets.