Research Papers

The Drawback of Insight: Detailed Explanations Can Reduce Agreement with XAI

Sabid Bin Habib Pias, Alicia Freel, Timothy Trammel +3

With the emergence of Artificial Intelligence (AI)-based decision-making, explanations help increase new technology adoption through enhanced trust and reliability. However, our experimental study challenges the notion that every user universally values explanations. We argue that the agreement with AI suggestions, whether accompanied by explanations or not, is influenced by individual differences in personality traits and the users' comfort with technology. We found that people with higher neuroticism and lower technological comfort showed more agreement with the recommendations without explanations. As more users become exposed to eXplainable AI (XAI) and AI-based systems, we argue that the XAI design should not provide explanations for users with high neuroticism and low technology comfort. Prioritizing user personalities in XAI systems will help users become better collaborators of AI systems.

Smart home systems are gaining popularity as homeowners strive to enhance their living and working environments while minimizing energy consumption. However, the adoption of artificial intelligence (AI)-enabled decision-making models in smart home systems faces challenges due to the complexity and black-box nature of these systems, leading to concerns about explainability, trust, transparency, accountability, and fairness. The emerging field of explainable artificial intelligence (XAI) addresses these issues by providing explanations for the models' decisions and actions. While state-of-the-art XAI methods are beneficial for AI developers and practitioners, they may not be easily understood by general users, particularly household members. This paper advocates for human-centered XAI methods, emphasizing the importance of delivering readily comprehensible explanations to enhance user satisfaction and drive the adoption of smart home systems. We review state-of-the-art XAI methods and prior studies focusing on human-centered explanations for general users in the context of smart home applications. Through experiments on two smart home application scenarios, we demonstrate that explanations generated by prominent XAI techniques might not be effective in helping users understand and make decisions. We thus argue for the necessity of a human-centric approach in representing explanations in smart home systems and highlight relevant human-computer interaction (HCI) methodologies, including user studies, prototyping, technology probes analysis, and heuristic evaluation, that can be employed to generate and present human-centered explanations to users.

Exploring Algorithmic Explainability: Generating Explainable AI Insights for Personalized Clinical Decision Support Focused on Cannabis Intoxication in Young Adults

Tongze Zhang, Tammy Chung, Anind Dey +1

This study explores the possibility of facilitating algorithmic decision-making by combining interpretable artificial intelligence (XAI) techniques with sensor data, with the aim of providing researchers and clinicians with personalized analyses of cannabis intoxication behavior. SHAP analyzes the importance and quantifies the impact of specific factors such as environmental noise or heart rate, enabling clinicians to pinpoint influential behaviors and environmental conditions. SkopeRules simplify the understanding of cannabis use for a specific activity or environmental use. Decision trees provide a clear visualization of how factors interact to influence cannabis consumption. Counterfactual models help identify key changes in behaviors or conditions that may alter cannabis use outcomes, to guide effective individualized intervention strategies. This multidimensional analytical approach not only unveils changes in behavioral and physiological states after cannabis use, such as frequent fluctuations in activity states, nontraditional sleep patterns, and specific use habits at different times and places, but also highlights the significance of individual differences in responses to cannabis use. These insights carry profound implications for clinicians seeking to gain a deeper understanding of the diverse needs of their patients and for tailoring precisely targeted intervention strategies. Furthermore, our findings highlight the pivotal role that XAI technologies could play in enhancing the transparency and interpretability of Clinical Decision Support Systems (CDSS), with a particular focus on substance misuse treatment. This research significantly contributes to ongoing initiatives aimed at advancing clinical practices that aim to prevent and reduce cannabis-related harms to health, positioning XAI as a supportive tool for clinicians and researchers alike.

Artificial intelligence supports healthcare professionals with predictive modeling, greatly transforming clinical decision-making. This study addresses the crucial need for fairness and explainability in AI applications within healthcare to ensure equitable outcomes across diverse patient demographics. By focusing on the predictive modeling of sepsis-related mortality, we propose a method that learns a performance-optimized predictive model and then employs the transfer learning process to produce a model with better fairness. Our method also introduces a novel permutation-based feature importance algorithm aiming at elucidating the contribution of each feature in enhancing fairness on predictions. Unlike existing explainability methods concentrating on explaining feature contribution to predictive performance, our proposed method uniquely bridges the gap in understanding how each feature contributes to fairness. This advancement is pivotal, given sepsis's significant mortality rate and its role in one-third of hospital deaths. Our method not only aids in identifying and mitigating biases within the predictive model but also fosters trust among healthcare stakeholders by improving the transparency and fairness of model predictions, thereby contributing to more equitable and trustworthy healthcare delivery.

Traditionally, assets are selected for inclusion in a portfolio (long or short) by human analysts. Teams of human portfolio managers (PMs) seek to weigh and balance these securities using optimisation methods and other portfolio construction processes. Often, human PMs consider human analyst recommendations against the backdrop of the analyst's recommendation track record and the applicability of the analyst to the recommendation they provide. Many firms regularly ask analysts to provide a "conviction" level on their recommendations. In the eyes of PMs, understanding a human analyst's track record has typically come down to basic spread sheet tabulation or, at best, a "virtual portfolio" paper trading book to keep track of results of recommendations. Analysts' conviction around their recommendations and their "paper trading" track record are two crucial workflow components between analysts and portfolio construction. Many human PMs may not even appreciate that they factor these data points into their decision-making logic. This chapter explores how Artificial Intelligence (AI) can be used to replicate these two steps and bridge the gap between AI data analytics and AI-based portfolio construction methods. This field of AI is referred to as Recommender Systems (RS). This chapter will further explore what metadata that RS systems functionally supply to downstream systems and their features.

Reactive Model Correction: Mitigating Harm to Task-Relevant Features via Conditional Bias Suppression

Dilyara Bareeva, Maximilian Dreyer, Frederik Pahde +2

Deep Neural Networks are prone to learning and relying on spurious correlations in the training data, which, for high-risk applications, can have fatal consequences. Various approaches to suppress model reliance on harmful features have been proposed that can be applied post-hoc without additional training. Whereas those methods can be applied with efficiency, they also tend to harm model performance by globally shifting the distribution of latent features. To mitigate unintended overcorrection of model behavior, we propose a reactive approach conditioned on model-derived knowledge and eXplainable Artificial Intelligence (XAI) insights. While the reactive approach can be applied to many post-hoc methods, we demonstrate the incorporation of reactivity in particular for P-ClArC (Projective Class Artifact Compensation), introducing a new method called R-ClArC (Reactive Class Artifact Compensation). Through rigorous experiments in controlled settings (FunnyBirds) and with a real-world dataset (ISIC2019), we show that introducing reactivity can minimize the detrimental effect of the applied correction while simultaneously ensuring low reliance on spurious features.

Due to the sensitive nature of medicine, it is particularly important and highly demanded that AI methods are explainable. This need has been recognised and there is great research interest in xAI solutions with medical applications. However, there is a lack of user-centred evaluation regarding the actual impact of the explanations. We evaluate attribute- and prototype-based explanations with the Proto-Caps model. This xAI model reasons the target classification with human-defined visual features of the target object in the form of scores and attribute-specific prototypes. The model thus provides a multimodal explanation that is intuitively understandable to humans thanks to predefined attributes. A user study involving six radiologists shows that the explanations are subjectivly perceived as helpful, as they reflect their decision-making process. The results of the model are considered a second opinion that radiologists can discuss using the model's explanations. However, it was shown that the inclusion and increased magnitude of model explanations objectively can increase confidence in the model's predictions when the model is incorrect. We can conclude that attribute scores and visual prototypes enhance confidence in the model. However, additional development and repeated user studies are needed to tailor the explanation to the respective use case.

Non-native English speakers (NNES) face challenges in digital workspace communication (e.g., emails, Slack messages), often inadvertently translating expressions from their native languages, which can lead to awkward or incorrect usage. Current AI-assisted writing tools are equipped with fluency enhancement and rewriting suggestions; however, NNES may struggle to grasp the subtleties among various expressions, making it challenging to choose the one that accurately reflects their intent. Such challenges are exacerbated in high-stake text-based communications, where the absence of non-verbal cues heightens the risk of misinterpretation. By leveraging the latest advancements in large language models (LLM) and word embeddings, we propose WordDecipher, an explainable AI-assisted writing tool to enhance digital workspace communication for NNES. WordDecipher not only identifies the perceived social intentions detected in users' writing, but also generates rewriting suggestions aligned with users' intended messages, either numerically or by inferring from users' writing in their native language. Then, WordDecipher provides an overview of nuances to help NNES make selections. Through a usage scenario, we demonstrate how WordDecipher can significantly enhance an NNES's ability to communicate her request, showcasing its potential to transform workspace communication for NNES.

Explainability algorithms aimed at interpreting decision-making AI systems usually consider balancing two critical dimensions: 1) \textit{faithfulness}, where explanations accurately reflect the model's inference process. 2) \textit{plausibility}, where explanations are consistent with domain experts. However, the question arises: do faithfulness and plausibility inherently conflict? In this study, through a comprehensive quantitative comparison between the explanations from the selected explainability methods and expert-level interpretations across three NLP tasks: sentiment analysis, intent detection, and topic labeling, we demonstrate that traditional perturbation-based methods Shapley value and LIME could attain greater faithfulness and plausibility. Our findings suggest that rather than optimizing for one dimension at the expense of the other, we could seek to optimize explainability algorithms with dual objectives to achieve high levels of accuracy and user accessibility in their explanations.

Red Teaming Models for Hyperspectral Image Analysis Using Explainable AI

Vladimir Zaigrajew, Hubert Baniecki, Lukasz Tulczyjew +4

Remote sensing (RS) applications in the space domain demand machine learning (ML) models that are reliable, robust, and quality-assured, making red teaming a vital approach for identifying and exposing potential flaws and biases. Since both fields advance independently, there is a notable gap in integrating red teaming strategies into RS. This paper introduces a methodology for examining ML models operating on hyperspectral images within the HYPERVIEW challenge, focusing on soil parameters' estimation. We use post-hoc explanation methods from the Explainable AI (XAI) domain to critically assess the best performing model that won the HYPERVIEW challenge and served as an inspiration for the model deployed on board the INTUITION-1 hyperspectral mission. Our approach effectively red teams the model by pinpointing and validating key shortcomings, constructing a model that achieves comparable performance using just 1% of the input features and a mere up to 5% performance loss. Additionally, we propose a novel way of visualizing explanations that integrate domain-specific information about hyperspectral bands (wavelengths) and data transformations to better suit interpreting models for hyperspectral image analysis.

The escalating integration of machine learning in high-stakes fields such as healthcare raises substantial concerns about model fairness. We propose an interpretable framework - Fairness-Aware Interpretable Modeling (FAIM), to improve model fairness without compromising performance, featuring an interactive interface to identify a "fairer" model from a set of high-performing models and promoting the integration of data-driven evidence and clinical expertise to enhance contextualized fairness. We demonstrated FAIM's value in reducing sex and race biases by predicting hospital admission with two real-world databases, MIMIC-IV-ED and SGH-ED. We show that for both datasets, FAIM models not only exhibited satisfactory discriminatory performance but also significantly mitigated biases as measured by well-established fairness metrics, outperforming commonly used bias-mitigation methods. Our approach demonstrates the feasibility of improving fairness without sacrificing performance and provides an a modeling mode that invites domain experts to engage, fostering a multidisciplinary effort toward tailored AI fairness.

In a recent paper, Erasmus et al. (2021) defend the idea that the ambiguity of the term "explanation" in explainable AI (XAI) can be solved by adopting any of four different extant accounts of explanation in the philosophy of science: the Deductive Nomological, Inductive Statistical, Causal Mechanical, and New Mechanist models. In this chapter, I show that the authors' claim that these accounts can be applied to deep neural networks as they would to any natural phenomenon is mistaken. I also provide a more general argument as to why the notion of explainability as it is currently used in the XAI literature bears little resemblance to the traditional concept of scientific explanation. It would be more fruitful to use the label "understandable AI" to avoid the confusion that surrounds the goal and purposes of XAI. In the second half of the chapter, I argue for a pragmatic conception of understanding that is better suited to play the central role attributed to explanation in XAI. Following Kuorikoski & Ylikoski (2015), the conditions of satisfaction for understanding an ML system are fleshed out in terms of an agent's success in using the system, in drawing correct inferences from it.