Research Papers

Fair Allocation through Selective Information Acquisition

William Cai, Johann Gaebler, Nikhil Garg +1

Public and private institutions must often allocate scare resources under uncertainty. Banks, for example, extend credit to loan applicants based in part on their estimated likelihood of repaying a loan. But when the quality of information differs across candidates (e.g., if some applicants lack traditional credit histories), common lending strategies can lead to disparities across groups. Here we consider a setting in which decision makers -- before allocating resources -- can choose to spend some of their limited budget further screening select individuals. We present a computationally efficient algorithm for deciding whom to screen that maximizes a standard measure of social welfare. Intuitively, decision makers should screen candidates on the margin, for whom the additional information could plausibly alter the allocation. We formalize this idea by showing the problem can be reduced to solving a series of linear programs. Both on synthetic and real-world datasets, this strategy improves utility, illustrating the value of targeted information acquisition in such decisions. Further, when there is social value for distributing resources to groups for whom we have a priori poor information -- like those without credit scores -- our approach can substantially improve the allocation of limited assets.

Explainable machine learning (ML) enables human learning from ML, human appeal of automated model decisions, regulatory compliance, and security audits of ML models. Explainable ML (i.e. explainable artificial intelligence or XAI) has been implemented in numerous open source and commercial packages and explainable ML is also an important, mandatory, or embedded aspect of commercial predictive modeling in industries like financial services. However, like many technologies, explainable ML can be misused, particularly as a faulty safeguard for harmful black-boxes, e.g. fairwashing or scaffolding, and for other malevolent purposes like stealing models and sensitive training data. To promote best-practice discussions for this already in-flight technology, this short text presents internal definitions and a few examples before covering the proposed guidelines. This text concludes with a seemingly natural argument for the use of interpretable models and explanatory, debugging, and disparate impact testing methods in life- or mission-critical ML systems.

A Network Intrusion Detection System (NIDS) monitors networks for cyber attacks and other unwanted activities. However, NIDS solutions often generate an overwhelming number of alerts daily, making it challenging for analysts to prioritize high-priority threats. While deep learning models promise to automate the prioritization of NIDS alerts, the lack of transparency in these models can undermine trust in their decision-making. This study highlights the critical need for explainable artificial intelligence (XAI) in NIDS alert classification to improve trust and interpretability. We employed a real-world NIDS alert dataset from Security Operations Center (SOC) of TalTech (Tallinn University Of Technology) in Estonia, developing a Long Short-Term Memory (LSTM) model to prioritize alerts. To explain the LSTM model's alert prioritization decisions, we implemented and compared four XAI methods: Local Interpretable Model-Agnostic Explanations (LIME), SHapley Additive exPlanations (SHAP), Integrated Gradients, and DeepLIFT. The quality of these XAI methods was assessed using a comprehensive framework that evaluated faithfulness, complexity, robustness, and reliability. Our results demonstrate that DeepLIFT consistently outperformed the other XAI methods, providing explanations with high faithfulness, low complexity, robust performance, and strong reliability. In collaboration with SOC analysts, we identified key features essential for effective alert classification. The strong alignment between these analyst-identified features and those obtained by the XAI methods validates their effectiveness and enhances the practical applicability of our approach.

Artificial intelligence (AI) has rapidly developed through advancements in computational power and the growth of massive datasets. However, this progress has also heightened challenges in interpreting the "black-box" nature of AI models. To address these concerns, eXplainable AI (XAI) has emerged with a focus on transparency and interpretability to enhance human understanding and trust in AI decision-making processes. In the context of multimodal data fusion and complex reasoning scenarios, the proposal of Multimodal eXplainable AI (MXAI) integrates multiple modalities for prediction and explanation tasks. Meanwhile, the advent of Large Language Models (LLMs) has led to remarkable breakthroughs in natural language processing, yet their complexity has further exacerbated the issue of MXAI. To gain key insights into the development of MXAI methods and provide crucial guidance for building more transparent, fair, and trustworthy AI systems, we review the MXAI methods from a historical perspective and categorize them across four eras: traditional machine learning, deep learning, discriminative foundation models, and generative LLMs. We also review evaluation metrics and datasets used in MXAI research, concluding with a discussion of future challenges and directions. A project related to this review has been created at https://github.com/ShilinSun/mxai_review.

Phishing has been a prevalent cyber threat that manipulates users into revealing sensitive private information through deceptive tactics, designed to masquerade as trustworthy entities. Over the years, proactively detection of phishing URLs (or websites) has been established as an widely-accepted defense approach. In literature, we often find supervised Machine Learning (ML) models with highly competitive performance for detecting phishing websites based on the extracted features from both phishing and benign (i.e., legitimate) websites. However, it is still unclear if these features or indicators are dependent on a particular dataset or they are generalized for overall phishing detection. In this paper, we delve deeper into this issue by analyzing two publicly available phishing URL datasets, where each dataset has its own set of unique and overlapping features related to URL string and website contents. We want to investigate if overlapping features are similar in nature across datasets and how does the model perform when trained on one dataset and tested on the other. We conduct practical experiments and leverage explainable AI (XAI) methods such as SHAP plots to provide insights into different features' contributions in case of phishing detection to answer our primary question, "Can features for phishing URL detection be trusted across diverse dataset?". Our case study experiment results show that features for phishing URL detection can often be dataset-dependent and thus may not be trusted across different datasets even though they share same set of feature behaviors.

Large Language Models (LLMs) are being adopted across a wide range of tasks, including decision-making processes in industries where bias in AI systems is a significant concern. Recent research indicates that LLMs can harbor implicit biases even when they pass explicit bias evaluations. Building upon the frameworks of the LLM Implicit Association Test (IAT) Bias and LLM Decision Bias, this study highlights that newer or larger language models do not automatically exhibit reduced bias; in some cases, they displayed higher bias scores than their predecessors, such as in Meta's Llama series and OpenAI's GPT models. This suggests that increasing model complexity without deliberate bias mitigation strategies can unintentionally amplify existing biases. The variability in bias scores within and across providers underscores the need for standardized evaluation metrics and benchmarks for bias assessment. The lack of consistency indicates that bias mitigation is not yet a universally prioritized goal in model development, which can lead to unfair or discriminatory outcomes. By broadening the detection of implicit bias, this research provides a more comprehensive understanding of the biases present in advanced models and underscores the critical importance of addressing these issues to ensure the development of fair and responsible AI systems.

The move toward artificial intelligence (AI)-native sixth-generation (6G) networks has put more emphasis on the importance of explainability and trustworthiness in network management operations, especially for mission-critical use-cases. Such desired trust transcends traditional post-hoc explainable AI (XAI) methods to using contextual explanations for guiding the learning process in an in-hoc way. This paper proposes a novel graph reinforcement learning (GRL) framework named TANGO which relies on a symbolic subsystem. It consists of a Bayesian-graph neural network (GNN) Explainer, whose outputs, in terms of edge/node importance and uncertainty, are periodically translated to a logical GRL reward function. This adjustment is accomplished through defined symbolic reasoning rules within a Reasoner. Considering a real-world testbed proof-of-concept (PoC), a gNodeB (gNB) radio resource allocation problem is formulated, which aims to minimize under- and over-provisioning of physical resource blocks (PRBs) while penalizing decisions emanating from the uncertain and less important edge-nodes relations. Our findings reveal that the proposed in-hoc explainability solution significantly expedites convergence compared to standard GRL baseline and other benchmarks in the deep reinforcement learning (DRL) domain. The experiment evaluates performance in AI, complexity, energy consumption, robustness, network, scalability, and explainability metrics. Specifically, the results show that TANGO achieves a noteworthy accuracy of 96.39% in terms of optimal PRB allocation in inference phase, outperforming the baseline by 1.22x.

This paper proposed a comprehensive mixed-methods framework with varied samples of older adults, including user experience, usability assessments, and in-depth interviews with the integration of Explainable Artificial Intelligence (XAI) methods. The experience of older adults' interaction with the Ehealth interface is collected through interviews and transformed into operatable databases whereas XAI methods are utilized to explain the collected interview results in this research work. The results show that XAI-infused e-health interfaces could play an important role in bridging the age-related digital divide by investigating elders' preferences when interacting with E-health interfaces. Furthermore, the study identifies important design factors, such as intuitive visualization and straightforward explanations, that are critical for creating efficient Human Computer Interaction (HCI) tools among older users. Furthermore, this study emphasizes the revolutionary potential of XAI in e-health interfaces for older users, emphasizing the importance of transparency and understandability in HCI-driven healthcare solutions. This study's findings have far-reaching implications for the design and development of user-centric e-health technologies, intending to increase the overall well-being of older adults.

Researchers, government bodies, and organizations have been repeatedly calling for a shift in the responsible AI community from general principles to tangible and operationalizable practices in mitigating the potential sociotechnical harms of AI. Frameworks like the NIST AI RMF embody an emerging consensus on recommended practices in operationalizing sociotechnical harm mitigation. However, private sector organizations currently lag far behind this emerging consensus. Implementation is sporadic and selective at best. At worst, it is ineffective and can risk serving as a misleading veneer of trustworthy processes, providing an appearance of legitimacy to substantively harmful practices. In this paper, we provide a foundation for a framework for evaluating where organizations sit relative to the emerging consensus on sociotechnical harm mitigation best practices: a flexible maturity model based on the NIST AI RMF.

A clinical artificial intelligence (AI) system is often validated on a held-out set of data which it has not been exposed to before (e.g., data from a different hospital with a distinct electronic health record system). This evaluation process is meant to mimic the deployment of an AI system on data in the wild; those which are currently unseen by the system yet are expected to be encountered in a clinical setting. However, when data in the wild differ from the held-out set of data, a phenomenon referred to as distribution shift, and lack ground-truth annotations, it becomes unclear the extent to which AI-based findings can be trusted on data in the wild. Here, we introduce SUDO, a framework for evaluating AI systems without ground-truth annotations. SUDO assigns temporary labels to data points in the wild and directly uses them to train distinct models, with the highest performing model indicative of the most likely label. Through experiments with AI systems developed for dermatology images, histopathology patches, and clinical reports, we show that SUDO can be a reliable proxy for model performance and thus identify unreliable predictions. We also demonstrate that SUDO informs the selection of models and allows for the previously out-of-reach assessment of algorithmic bias for data in the wild without ground-truth annotations. The ability to triage unreliable predictions for further inspection and assess the algorithmic bias of AI systems can improve the integrity of research findings and contribute to the deployment of ethical AI systems in medicine.

The interplay between artificial intelligence (AI) and psychology, particularly in personality assessment, represents an important emerging area of research. Accurate personality trait estimation is crucial not only for enhancing personalization in human-computer interaction but also for a wide variety of applications ranging from mental health to education. This paper analyzes the capability of a generic chatbot, ChatGPT, to effectively infer personality traits from short texts. We report the results of a comprehensive user study featuring texts written in Czech by a representative population sample of 155 participants. Their self-assessments based on the Big Five Inventory (BFI) questionnaire serve as the ground truth. We compare the personality trait estimations made by ChatGPT against those by human raters and report ChatGPT's competitive performance in inferring personality traits from text. We also uncover a 'positivity bias' in ChatGPT's assessments across all personality dimensions and explore the impact of prompt composition on accuracy. This work contributes to the understanding of AI capabilities in psychological assessment, highlighting both the potential and limitations of using large language models for personality inference. Our research underscores the importance of responsible AI development, considering ethical implications such as privacy, consent, autonomy, and bias in AI applications.

The evaluation of the fidelity of eXplainable Artificial Intelligence (XAI) methods to their underlying models is a challenging task, primarily due to the absence of a ground truth for explanations. However, assessing fidelity is a necessary step for ensuring a correct XAI methodology. In this study, we conduct a fair and objective comparison of the current state-of-the-art XAI methods by introducing three novel image datasets with reliable ground truth for explanations. The primary objective of this comparison is to identify methods with low fidelity and eliminate them from further research, thereby promoting the development of more trustworthy and effective XAI techniques. Our results demonstrate that XAI methods based on the backpropagation of output information to input yield higher accuracy and reliability compared to methods relying on sensitivity analysis or Class Activation Maps (CAM). However, the backpropagation method tends to generate more noisy saliency maps. These findings have significant implications for the advancement of XAI methods, enabling the elimination of erroneous explanations and fostering the development of more robust and reliable XAI.