Research Papers

Quantifying Explainability of Saliency Methods in Deep Neural Networks with a Synthetic Dataset

Erico Tjoa, Cuntai Guan

Post-hoc analysis is a popular category in eXplainable artificial intelligence (XAI) study. In particular, methods that generate heatmaps have been used to explain the deep neural network (DNN), a black-box model. Heatmaps can be appealing due to the intuitive and visual ways to understand them but assessing their qualities might not be straightforward. Different ways to assess heatmaps' quality have their own merits and shortcomings. This paper introduces a synthetic dataset that can be generated adhoc along with the ground-truth heatmaps for more objective quantitative assessment. Each sample data is an image of a cell with easily recognized features that are distinguished from localization ground-truth mask, hence facilitating a more transparent assessment of different XAI methods. Comparison and recommendations are made, shortcomings are clarified along with suggestions for future research directions to handle the finer details of select post-hoc analysis methods. Furthermore, mabCAM is introduced as the heatmap generation method compatible with our ground-truth heatmaps. The framework is easily generalizable and uses only standard deep learning components.

Gender classification algorithms have important applications in many domains today such as demographic research, law enforcement, as well as human-computer interaction. Recent research showed that algorithms trained on biased benchmark databases could result in algorithmic bias. However, to date, little research has been carried out on gender classification algorithms' bias towards gender minorities subgroups, such as the LGBTQ and the non-binary population, who have distinct characteristics in gender expression. In this paper, we began by conducting surveys on existing benchmark databases for facial recognition and gender classification tasks. We discovered that the current benchmark databases lack representation of gender minority subgroups. We worked on extending the current binary gender classifier to include a non-binary gender class. We did that by assembling two new facial image databases: 1) a racially balanced inclusive database with a subset of LGBTQ population 2) an inclusive-gender database that consists of people with non-binary gender. We worked to increase classification accuracy and mitigate algorithmic biases on our baseline model trained on the augmented benchmark database. Our ensemble model has achieved an overall accuracy score of 90.39%, which is a 38.72% increase from the baseline binary gender classifier trained on Adience. While this is an initial attempt towards mitigating bias in gender classification, more work is needed in modeling gender as a continuum by assembling more inclusive databases.

Directions for Explainable Knowledge-Enabled Systems

Shruthi Chari, Daniel M. Gruen, Oshani Seneviratne +1

Interest in the field of Explainable Artificial Intelligence has been growing for decades and has accelerated recently. As Artificial Intelligence models have become more complex, and often more opaque, with the incorporation of complex machine learning techniques, explainability has become more critical. Recently, researchers have been investigating and tackling explainability with a user-centric focus, looking for explanations to consider trustworthiness, comprehensibility, explicit provenance, and context-awareness. In this chapter, we leverage our survey of explanation literature in Artificial Intelligence and closely related fields and use these past efforts to generate a set of explanation types that we feel reflect the expanded needs of explanation for today's artificial intelligence applications. We define each type and provide an example question that would motivate the need for this style of explanation. We believe this set of explanation types will help future system designers in their generation and prioritization of requirements and further help generate explanations that are better aligned to users' and situational needs.

Explainable Active Learning (XAL): An Empirical Study of How Local Explanations Impact Annotator Experience

Bhavya Ghai, Q. Vera Liao, Yunfeng Zhang +2

The wide adoption of Machine Learning technologies has created a rapidly growing demand for people who can train ML models. Some advocated the term "machine teacher" to refer to the role of people who inject domain knowledge into ML models. One promising learning paradigm is Active Learning (AL), by which the model intelligently selects instances to query the machine teacher for labels. However, in current AL settings, the human-AI interface remains minimal and opaque. We begin considering AI explanations as a core element of the human-AI interface for teaching machines. When a human student learns, it is a common pattern to present one's own reasoning and solicit feedback from the teacher. When a ML model learns and still makes mistakes, the human teacher should be able to understand the reasoning underlying the mistakes. When the model matures, the machine teacher should be able to recognize its progress in order to trust and feel confident about their teaching outcome. Toward this vision, we propose a novel paradigm of explainable active learning (XAL), by introducing techniques from the recently surging field of explainable AI (XAI) into an AL setting. We conducted an empirical study comparing the model learning outcomes, feedback content and experience with XAL, to that of traditional AL and coactive learning (providing the model's prediction without the explanation). Our study shows benefits of AI explanation as interfaces for machine teaching--supporting trust calibration and enabling rich forms of teaching feedback, and potential drawbacks--anchoring effect with the model judgment and cognitive workload. Our study also reveals important individual factors that mediate a machine teacher's reception to AI explanations, including task knowledge, AI experience and need for cognition. By reflecting on the results, we suggest future directions and design implications for XAL.

In today's legal environment, lawsuits and regulatory investigations require companies to embark upon increasingly intensive data-focused engagements to identify, collect and analyze large quantities of data. When documents are staged for review the process can require companies to dedicate an extraordinary level of resources, both with respect to human resources, but also with respect to the use of technology-based techniques to intelligently sift through data. For several years, attorneys have been using a variety of tools to conduct this exercise, and most recently, they are accepting the use of machine learning techniques like text classification to efficiently cull massive volumes of data to identify responsive documents for use in these matters. In recent years, a group of AI and Machine Learning researchers have been actively researching Explainable AI. In an explainable AI system, actions or decisions are human understandable. In typical legal `document review' scenarios, a document can be identified as responsive, as long as one or more of the text snippets in a document are deemed responsive. In these scenarios, if predictive coding can be used to locate these responsive snippets, then attorneys could easily evaluate the model's document classification decision. When deployed with defined and explainable results, predictive coding can drastically enhance the overall quality and speed of the document review process by reducing the time it takes to review documents. The authors of this paper propose the concept of explainable predictive coding and simple explainable predictive coding methods to locate responsive snippets within responsive documents. We also report our preliminary experimental results using the data from an actual legal matter that entailed this type of document review.

Explaining Explanations to Society

Leilani H. Gilpin, Cecilia Testart, Nathaniel Fruchter +1

There is a disconnect between explanatory artificial intelligence (XAI) methods and the types of explanations that are useful for and demanded by society (policy makers, government officials, etc.) Questions that experts in artificial intelligence (AI) ask opaque systems provide inside explanations, focused on debugging, reliability, and validation. These are different from those that society will ask of these systems to build trust and confidence in their decisions. Although explanatory AI systems can answer many questions that experts desire, they often don't explain why they made decisions in a way that is precise (true to the model) and understandable to humans. These outside explanations can be used to build trust, comply with regulatory and policy changes, and act as external validation. In this paper, we focus on XAI methods for deep neural networks (DNNs) because of DNNs' use in decision-making and inherent opacity. We explore the types of questions that explanatory DNN systems can answer and discuss challenges in building explanatory systems that provide outside explanations for societal requirements and benefit.

Assessing and understanding intelligent agents is a difficult task for users that lack an AI background. A relatively new area, called "Explainable AI," is emerging to help address this problem, but little is known about how users would forage through information an explanation system might offer. To inform the development of Explainable AI systems, we conducted a formative study, using the lens of Information Foraging Theory, into how experienced users foraged in the domain of StarCraft to assess an agent. Our results showed that participants faced difficult foraging problems. These foraging problems caused participants to entirely miss events that were important to them, reluctantly choose to ignore actions they did not want to ignore, and bear high cognitive, navigation, and information costs to access the information they needed.

As artificial intelligence (AI) advances in reasoning capabilities, most recently with the emergence of Large Reasoning Models (LRMs), understanding how children conceptualize AI's reasoning processes becomes critical for fostering AI literacy. While one of the "Five Big Ideas" in AI education highlights reasoning algorithms as central to AI decision-making, less is known about children's mental models in this area. Through a two-phase approach, consisting of a co-design session with 8 children followed by a field study with 106 children (grades 3-8), we identified three models of AI reasoning: Deductive, Inductive, and Inherent. Our findings reveal that younger children (grades 3-5) often attribute AI's reasoning to inherent intelligence, while older children (grades 6-8) recognize AI as a pattern recognizer. We highlight three tensions that surfaced in children's understanding of AI reasoning and conclude with implications for scaffolding AI curricula and designing explainable AI tools.

Large language models (LLMs) could be valuable personal AI agents across various domains, provided they can precisely follow user instructions. However, recent studies have shown significant limitations in LLMs' instruction-following capabilities, raising concerns about their reliability in high-stakes applications. Accurately estimating LLMs' uncertainty in adhering to instructions is critical to mitigating deployment risks. We present, to our knowledge, the first systematic evaluation of the uncertainty estimation abilities of LLMs in the context of instruction-following. Our study identifies key challenges with existing instruction-following benchmarks, where multiple factors are entangled with uncertainty stems from instruction-following, complicating the isolation and comparison across methods and models. To address these issues, we introduce a controlled evaluation setup with two benchmark versions of data, enabling a comprehensive comparison of uncertainty estimation methods under various conditions. Our findings show that existing uncertainty methods struggle, particularly when models make subtle errors in instruction following. While internal model states provide some improvement, they remain inadequate in more complex scenarios. The insights from our controlled evaluation setups provide a crucial understanding of LLMs' limitations and potential for uncertainty estimation in instruction-following tasks, paving the way for more trustworthy AI agents.

COMPL-AI Framework: A Technical Interpretation and LLM Benchmarking Suite for the EU Artificial Intelligence Act

Philipp Guldimann, Alexander Spiridonov, Robin Staab +9

The EU's Artificial Intelligence Act (AI Act) is a significant step towards responsible AI development, but lacks clear technical interpretation, making it difficult to assess models' compliance. This work presents COMPL-AI, a comprehensive framework consisting of (i) the first technical interpretation of the EU AI Act, translating its broad regulatory requirements into measurable technical requirements, with the focus on large language models (LLMs), and (ii) an open-source Act-centered benchmarking suite, based on thorough surveying and implementation of state-of-the-art LLM benchmarks. By evaluating 12 prominent LLMs in the context of COMPL-AI, we reveal shortcomings in existing models and benchmarks, particularly in areas like robustness, safety, diversity, and fairness. This work highlights the need for a shift in focus towards these aspects, encouraging balanced development of LLMs and more comprehensive regulation-aligned benchmarks. Simultaneously, COMPL-AI for the first time demonstrates the possibilities and difficulties of bringing the Act's obligations to a more concrete, technical level. As such, our work can serve as a useful first step towards having actionable recommendations for model providers, and contributes to ongoing efforts of the EU to enable application of the Act, such as the drafting of the GPAI Code of Practice.

Utilizing potent representations of the large vision-language models (VLMs) to accomplish various downstream tasks has attracted increasing attention. Within this research field, soft prompt learning has become a representative approach for efficiently adapting VLMs such as CLIP, to tasks like image classification. However, most existing prompt learning methods learn text tokens that are unexplainable, which cannot satisfy the stringent interpretability requirements of Explainable Artificial Intelligence (XAI) in high-stakes scenarios like healthcare. To address this issue, we propose a novel explainable prompt learning framework that leverages medical knowledge by aligning the semantics of images, learnable prompts, and clinical concept-driven prompts at multiple granularities. Moreover, our framework addresses the lack of valuable concept annotations by eliciting knowledge from large language models and offers both visual and textual explanations for the prompts. Extensive experiments and explainability analyses conducted on various datasets, with and without concept labels, demonstrate that our method simultaneously achieves superior diagnostic performance, flexibility, and interpretability, shedding light on the effectiveness of foundation models in facilitating XAI. The code will be made publically available.

This review paper provides an integrated perspective of Explainable Artificial Intelligence techniques applied to Brain-Computer Interfaces. BCIs use predictive models to interpret brain signals for various high-stake applications. However, achieving explainability in these complex models is challenging as it compromises accuracy. The field of XAI has emerged to address the need for explainability across various stakeholders, but there is a lack of an integrated perspective in XAI for BCI (XAI4BCI) literature. It is necessary to differentiate key concepts like explainability, interpretability, and understanding in this context and formulate a comprehensive framework. To understand the need of XAI for BCI, we pose six key research questions for a systematic review and meta-analysis, encompassing its purposes, applications, usability, and technical feasibility. We employ the PRISMA methodology -- preferred reporting items for systematic reviews and meta-analyses to review (n=1246) and analyze (n=84) studies published in 2015 and onwards for key insights. The results highlight that current research primarily focuses on interpretability for developers and researchers, aiming to justify outcomes and enhance model performance. We discuss the unique approaches, advantages, and limitations of XAI4BCI from the literature. We draw insights from philosophy, psychology, and social sciences. We propose a design space for XAI4BCI, considering the evolving need to visualize and investigate predictive model outcomes customised for various stakeholders in the BCI development and deployment lifecycle. This paper is the first to focus solely on reviewing XAI4BCI research articles. This systematic review and meta-analysis findings with the proposed design space prompt important discussions on establishing standards for BCI explanations, highlighting current limitations, and guiding the future of XAI in BCI.