Research Papers

The extensive utilization of biometric authentication systems have emanated attackers / imposters to forge user identity based on morphed images. In this attack, a synthetic image is produced and merged with genuine. Next, the resultant image is user for authentication. Numerous deep neural convolutional architectures have been proposed in literature for face Morphing Attack Detection (MADs) to prevent such attacks and lessen the risks associated with them. Although, deep learning models achieved optimal results in terms of performance, it is difficult to understand and analyse these networks since they are black box/opaque in nature. As a consequence, incorrect judgments may be made. There is, however, a dearth of literature that explains decision-making methods of black box deep learning models for biometric Presentation Attack Detection (PADs) or MADs that can aid the biometric community to have trust in deep learning-based biometric systems for identification and authentication in various security applications such as border control, criminal database establishment etc. In this work, we present a novel visual explanation approach named Ensemble XAI integrating Saliency maps, Class Activation Maps (CAM) and Gradient-CAM (Grad-CAM) to provide a more comprehensive visual explanation for a deep learning prognostic model (EfficientNet-B1) that we have employed to predict whether the input presented to a biometric authentication system is morphed or genuine. The experimentations have been performed on three publicly available datasets namely Face Research Lab London Set, Wide Multi-Channel Presentation Attack (WMCA), and Makeup Induced Face Spoofing (MIFS). The experimental evaluations affirms that the resultant visual explanations highlight more fine-grained details of image features/areas focused by EfficientNet-B1 to reach decisions along with appropriate reasoning.

Explaining Classifications to Non Experts: An XAI User Study of Post Hoc Explanations for a Classifier When People Lack Expertise

Courtney Ford, Mark T Keane

Very few eXplainable AI (XAI) studies consider how users understanding of explanations might change depending on whether they know more or less about the to be explained domain (i.e., whether they differ in their expertise). Yet, expertise is a critical facet of most high stakes, human decision making (e.g., understanding how a trainee doctor differs from an experienced consultant). Accordingly, this paper reports a novel, user study (N=96) on how peoples expertise in a domain affects their understanding of post-hoc explanations by example for a deep-learning, black box classifier. The results show that peoples understanding of explanations for correct and incorrect classifications changes dramatically, on several dimensions (e.g., response times, perceptions of correctness and helpfulness), when the image-based domain considered is familiar (i.e., MNIST) as opposed to unfamiliar (i.e., Kannada MNIST). The wider implications of these new findings for XAI strategies are discussed.

We identify the task of measuring data to quantitatively characterize the composition of machine learning data and datasets. Similar to an object's height, width, and volume, data measurements quantify different attributes of data along common dimensions that support comparison. Several lines of research have proposed what we refer to as measurements, with differing terminology; we bring some of this work together, particularly in fields of computer vision and language, and build from it to motivate measuring data as a critical component of responsible AI development. Measuring data aids in systematically building and analyzing machine learning (ML) data towards specific goals and gaining better control of what modern ML systems will learn. We conclude with a discussion of the many avenues of future work, the limitations of data measurements, and how to leverage these measurement approaches in research and practice.

While certain industrial sectors (e.g., aviation) have a long history of mandatory incident reporting complete with analytical findings, the practice of artificial intelligence (AI) safety benefits from no such mandate and thus analyses must be performed on publicly known ``open source'' AI incidents. Although the exact causes of AI incidents are seldom known by outsiders, this work demonstrates how to apply expert knowledge on the population of incidents in the AI Incident Database (AIID) to infer the potential and likely technical causative factors that contribute to reported failures and harms. We present early work on a taxonomic system that covers a cascade of interrelated incident factors, from system goals (nearly always known) to methods / technologies (knowable in many cases) and technical failure causes (subject to expert analysis) of the implicated systems. We pair this ontology structure with a comprehensive classification workflow that leverages expert knowledge and community feedback, resulting in taxonomic annotations grounded by incident data and human expertise.

Parametrized Quantum Circuits (PQCs) enable a novel method for machine learning (ML). However, from a computational point of view they present a challenge to existing eXplainable AI (xAI) methods. On the one hand, measurements on quantum circuits introduce probabilistic errors which impact the convergence of these methods. On the other hand, the phase space of a quantum circuit expands exponentially with the number of qubits, complicating efforts to execute xAI methods in polynomial time. In this paper we will discuss the performance of established xAI methods, such as Baseline SHAP and Integrated Gradients. Using the internal mechanics of PQCs we study ways to speed up their computation.

Artificial intelligence-empowred Real-Time Bidding (AIRTB) is regarded as one of the most enabling technologies for online advertising. It has attracted significant research attention from diverse fields such as pattern recognition, game theory and mechanism design. Despite of its remarkable development and deployment, the AIRTB system can sometimes harm the interest of its participants (e.g., depleting the advertisers' budget with various kinds of fraud). As such, building trustworthy AIRTB auctioning systems has emerged as an important direction of research in this field in recent years. Due to the highly interdisciplinary nature of this field and a lack of a comprehensive survey, it is a challenge for researchers to enter this field and contribute towards building trustworthy AIRTB technologies. This paper bridges this important gap in trustworthy AIRTB literature. We start by analysing the key concerns of various AIRTB stakeholders and identify three main dimensions of trust building in AIRTB, namely security, robustness and fairness. For each of these dimensions, we propose a unique taxonomy of the state of the art, trace the root causes of possible breakdown of trust, and discuss the necessity of the given dimension. This is followed by a comprehensive review of existing strategies for fulfilling the requirements of each trust dimension. In addition, we discuss the promising future directions of research essential towards building trustworthy AIRTB systems to benefit the field of online advertising.

There is still a significant gap between expectations and the successful adoption of AI to innovate and improve businesses. Due to the emergence of deep learning, AI adoption is more complex as it often incorporates big data and the internet of things, affecting data privacy. Existing frameworks have identified the need to focus on human-centered design, combining technical and business/organizational perspectives. However, trust remains a critical issue that needs to be designed from the beginning. The proposed framework expands from the human-centered design approach, emphasizing and maintaining the trust that underpins the process. This paper proposes a theoretical framework for responsible artificial intelligence (AI) implementation. The proposed framework emphasizes a synergistic business technology approach for the agile co-creation process. The aim is to streamline the adoption process of AI to innovate and improve business by involving all stakeholders throughout the project so that the AI technology is designed, developed, and deployed in conjunction with people and not in isolation. The framework presents a fresh viewpoint on responsible AI implementation based on analytical literature review, conceptual framework design, and practitioners' mediating expertise. The framework emphasizes establishing and maintaining trust throughout the human-centered design and agile development of AI. This human-centered approach is aligned with and enabled by the privacy by design principle. The creators of the technology and the end-users are working together to tailor the AI solution specifically for the business requirements and human characteristics. An illustrative case study on adopting AI for assisting planning in a hospital will demonstrate that the proposed framework applies to real-life applications.

Explaining the behaviour of intelligent systems will get increasingly and perhaps intractably challenging as models grow in size and complexity. We may not be able to expect an explanation for every prediction made by a brain-scale model, nor can we expect explanations to remain objective or apolitical. Our functionalist understanding of these models is of less advantage than we might assume. Models precede explanations, and can be useful even when both model and explanation are incorrect. Explainability may never win the race against complexity, but this is less problematic than it seems.

This study investigates the impact of machine learning models on the generation of counterfactual explanations by conducting a benchmark evaluation over three different types of models: a decision tree (fully transparent, interpretable, white-box model), a random forest (semi-interpretable, grey-box model), and a neural network (fully opaque, black-box model). We tested the counterfactual generation process using four algorithms (DiCE, WatcherCF, prototype, and GrowingSpheresCF) in the literature in 25 different datasets. Our findings indicate that: (1) Different machine learning models have little impact on the generation of counterfactual explanations; (2) Counterfactual algorithms based uniquely on proximity loss functions are not actionable and will not provide meaningful explanations; (3) One cannot have meaningful evaluation results without guaranteeing plausibility in the counterfactual generation. Algorithms that do not consider plausibility in their internal mechanisms will lead to biased and unreliable conclusions if evaluated with the current state-of-the-art metrics; (4) A counterfactual inspection analysis is strongly recommended to ensure a robust examination of counterfactual explanations and the potential identification of biases.

We investigate the performance of a state-of-the art (SoTA) architecture T5 (available on the SuperGLUE) and compare with it 3 other previous SoTA architectures across 5 different tasks from 2 relatively diverse datasets. The datasets are diverse in terms of the number and types of tasks they have. To improve performance, we augment the training data by using an autoregressive model. We achieve near-SoTA results on a couple of the tasks - macro F1 scores of 81.66% for task A of the OLID 2019 dataset and 82.54% for task A of the hate speech and offensive content (HASOC) 2021 dataset, where SoTA are 82.9% and 83.05%, respectively. We perform error analysis and explain why one of the models (Bi-LSTM) makes the predictions it does by using a publicly available algorithm: Integrated Gradient (IG). This is because explainable artificial intelligence (XAI) is essential for earning the trust of users. The main contributions of this work are the implementation method of T5, which is discussed; the data augmentation using a new conversational AI model checkpoint, which brought performance improvements; and the revelation on the shortcomings of HASOC 2021 dataset. It reveals the difficulties of poor data annotation by using a small set of examples where the T5 model made the correct predictions, even when the ground truth of the test set were incorrect (in our opinion). We also provide our model checkpoints on the HuggingFace hub1 to foster transparency.

Fairness Implications of Encoding Protected Categorical Attributes

Carlos Mougan, Jose M. Alvarez, Salvatore Ruggieri +1

Past research has demonstrated that the explicit use of protected attributes in machine learning can improve both performance and fairness. Many machine learning algorithms, however, cannot directly process categorical attributes, such as country of birth or ethnicity. Because protected attributes frequently are categorical, they must be encoded as features that can be input to a chosen machine learning algorithm, e.g.\ support vector machines, gradient boosting decision trees or linear models. Thereby, encoding methods influence how and what the machine learning algorithm will learn, affecting model performance and fairness. This work compares the accuracy and fairness implications of the two most well-known encoding methods: \emph{one-hot encoding} and \emph{target encoding}. We distinguish between two types of induced bias that may arise from these encoding methods and may lead to unfair models. The first type, \textit{irreducible bias}, is due to direct group category discrimination, and the second type, \textit{reducible bias}, is due to the large variance in statistically underrepresented groups. We investigate the interaction between categorical encodings and target encoding regularization methods that reduce unfairness. Furthermore, we consider the problem of intersectional unfairness that may arise when machine learning best practices improve performance measures by encoding several categorical attributes into a high-cardinality feature.

Explanation as a process: user-centric construction of multi-level and multi-modal explanations

Bettina Finzel, David E. Tafler, Stephan Scheele +1

In the last years, XAI research has mainly been concerned with developing new technical approaches to explain deep learning models. Just recent research has started to acknowledge the need to tailor explanations to different contexts and requirements of stakeholders. Explanations must not only suit developers of models, but also domain experts as well as end users. Thus, in order to satisfy different stakeholders, explanation methods need to be combined. While multi-modal explanations have been used to make model predictions more transparent, less research has focused on treating explanation as a process, where users can ask for information according to the level of understanding gained at a certain point in time. Consequently, an opportunity to explore explanations on different levels of abstraction should be provided besides multi-modal explanations. We present a process-based approach that combines multi-level and multi-modal explanations. The user can ask for textual explanations or visualizations through conversational interaction in a drill-down manner. We use Inductive Logic Programming, an interpretable machine learning approach, to learn a comprehensible model. Further, we present an algorithm that creates an explanatory tree for each example for which a classifier decision is to be explained. The explanatory tree can be navigated by the user to get answers of different levels of detail. We provide a proof-of-concept implementation for concepts induced from a semantic net about living beings.