Research Papers

Measuring Information in Text Explanations

Zining Zhu, Frank Rudzicz

Text-based explanation is a particularly promising approach in explainable AI, but the evaluation of text explanations is method-dependent. We argue that placing the explanations on an information-theoretic framework could unify the evaluations of two popular text explanation methods: rationale and natural language explanations (NLE). This framework considers the post-hoc text pipeline as a series of communication channels, which we refer to as ``explanation channels''. We quantify the information flow through these channels, thereby facilitating the assessment of explanation characteristics. We set up tools for quantifying two information scores: relevance and informativeness. We illustrate what our proposed information scores measure by comparing them against some traditional evaluation metrics. Our information-theoretic scores reveal some unique observations about the underlying mechanisms of two representative text explanations. For example, the NLEs trade-off slightly between transmitting the input-related information and the target-related information, whereas the rationales do not exhibit such a trade-off mechanism. Our work contributes to the ongoing efforts in establishing rigorous and standardized evaluation criteria in the rapidly evolving field of explainable AI.

Towards Attributions of Input Variables in a Coalition

Xinhao Zheng, Huiqi Deng, Quanshi Zhang

This paper focuses on the fundamental challenge of partitioning input variables in attribution methods for Explainable AI, particularly in Shapley value-based approaches. Previous methods always compute attributions given a predefined partition but lack theoretical guidance on how to form meaningful variable partitions. We identify that attribution conflicts arise when the attribution of a coalition differs from the sum of its individual variables' attributions. To address this, we analyze the numerical effects of AND-OR interactions in AI models and extend the Shapley value to a new attribution metric for variable coalitions. Our theoretical findings reveal that specific interactions cause attribution conflicts, and we propose three metrics to evaluate coalition faithfulness. Experiments on synthetic data, NLP, image classification, and the game of Go validate our approach, demonstrating consistency with human intuition and practical applicability.

The field of eXplainable Artificial Intelligence faces challenges due to the absence of a widely accepted taxonomy that facilitates the quantitative evaluation of explainability in Machine Learning algorithms. In this paper, we propose a novel taxonomy that addresses the current gap in the literature by providing a clear and unambiguous understanding of the key concepts and relationships in XAI. Our approach is rooted in a systematic analysis of existing definitions and frameworks, with a focus on transparency, interpretability, completeness, complexity and understandability as essential dimensions of explainability. This comprehensive taxonomy aims to establish a shared vocabulary for future research. To demonstrate the utility of our proposed taxonomy, we examine a case study of a Recommender System designed to curate and recommend the most suitable online resources from MERLOT. By employing the SHAP package, we quantify and enhance the explainability of the RS within the context of our newly developed taxonomy.

Adaptive Adversarial Training Does Not Increase Recourse Costs

Ian Hardy, Jayanth Yetukuri, Yang Liu

Recent work has connected adversarial attack methods and algorithmic recourse methods: both seek minimal changes to an input instance which alter a model's classification decision. It has been shown that traditional adversarial training, which seeks to minimize a classifier's susceptibility to malicious perturbations, increases the cost of generated recourse; with larger adversarial training radii correlating with higher recourse costs. From the perspective of algorithmic recourse, however, the appropriate adversarial training radius has always been unknown. Another recent line of work has motivated adversarial training with adaptive training radii to address the issue of instance-wise variable adversarial vulnerability, showing success in domains with unknown attack radii. This work studies the effects of adaptive adversarial training on algorithmic recourse costs. We establish that the improvements in model robustness induced by adaptive adversarial training show little effect on algorithmic recourse costs, providing a potential avenue for affordable robustness in domains where recoursability is critical.

Algorithm fairness has become a central problem for the broad adoption of artificial intelligence. Although the past decade has witnessed an explosion of excellent work studying algorithm biases, achieving fairness in real-world AI production systems has remained a challenging task. Most existing works fail to excel in practical applications since either they have conflicting measurement techniques and/ or heavy assumptions, or require code-access of the production models, whereas real systems demand an easy-to-implement measurement framework and a systematic way to correct the detected sources of bias. In this paper, we leverage recent advances in causal inference and interpretable machine learning to present an algorithm-agnostic framework (MIIF) to Measure, Interpret, and Improve the Fairness of an algorithmic decision. We measure the algorithm bias using randomized experiments, which enables the simultaneous measurement of disparate treatment, disparate impact, and economic value. Furthermore, using modern interpretability techniques, we develop an explainable machine learning model which accurately interprets and distills the beliefs of a blackbox algorithm. Altogether, these techniques create a simple and powerful toolset for studying algorithm fairness, especially for understanding the cost of fairness in practical applications like e-commerce and targeted advertising, where industry A/B testing is already abundant.

Declarative Reasoning on Explanations Using Constraint Logic Programming

Laura State, Salvatore Ruggieri, Franco Turini

Explaining opaque Machine Learning (ML) models is an increasingly relevant problem. Current explanation in AI (XAI) methods suffer several shortcomings, among others an insufficient incorporation of background knowledge, and a lack of abstraction and interactivity with the user. We propose REASONX, an explanation method based on Constraint Logic Programming (CLP). REASONX can provide declarative, interactive explanations for decision trees, which can be the ML models under analysis or global/local surrogate models of any black-box model. Users can express background or common sense knowledge using linear constraints and MILP optimization over features of factual and contrastive instances, and interact with the answer constraints at different levels of abstraction through constraint projection. We present here the architecture of REASONX, which consists of a Python layer, closer to the user, and a CLP layer. REASONX's core execution engine is a Prolog meta-program with declarative semantics in terms of logic theories.

How to make artificial intelligence (AI) systems safe and aligned with human values is an open research question. Proposed solutions tend toward relying on human intervention in uncertain situations, learning human values and intentions through training or observation, providing off-switches, implementing isolation or simulation environments, or extrapolating what people would want if they had more knowledge and more time to think. Law-based approaches--such as inspired by Isaac Asimov--have not been well regarded. This paper makes a case that effective legal systems are the best way to address AI safety. Law is defined as any rules that codify prohibitions and prescriptions applicable to particular agents in specified domains/contexts and includes processes for enacting, managing, enforcing, and litigating such rules.

Artificial intelligence (AI) reasoning and explainable AI (XAI) tasks have gained popularity recently, enabling users to explain the predictions or decision processes of AI models. This paper introduces Forest Monkey (FM), a toolkit designed to reason the outputs of any AI-based defect detection and/or classification model with data explainability. Implemented as a Python package, FM takes input in the form of dataset folder paths (including original images, ground truth labels, and predicted labels) and provides a set of charts and a text file to illustrate the reasoning results and suggest possible improvements. The FM toolkit consists of processes such as feature extraction from predictions to reasoning targets, feature extraction from images to defect characteristics, and a decision tree-based AI-Reasoner. Additionally, this paper investigates the time performance of the FM toolkit when applied to four AI models with different datasets. Lastly, a tutorial is provided to guide users in performing reasoning tasks using the FM toolkit.

We often see the term explainable in the titles of papers that describe applications based on artificial intelligence (AI). However, the literature in explainable artificial intelligence (XAI) indicates that explanations in XAI are application- and domain-specific, hence requiring evaluation whenever they are employed to explain a model that makes decisions for a specific application problem. Additionally, the literature reveals that the performance of post-hoc methods, particularly feature attribution methods, varies substantially hinting that they do not represent a solution to AI explainability. Therefore, when using XAI methods, the quality and suitability of their information outputs should be evaluated within the specific application. For these reasons, we used a scoping review methodology to investigate papers that apply AI models and adopt methods to generate post-hoc explanations while referring to said models as explainable. This paper investigates whether the term explainable model is adopted by authors under the assumption that incorporating a post-hoc XAI method suffices to characterize a model as explainable. To inspect this problem, our review analyzes whether these papers conducted evaluations. We found that 81% of the application papers that refer to their approaches as an explainable model do not conduct any form of evaluation on the XAI method they used.

Radio galaxy morphological classification is one of the critical steps when producing source catalogues for large-scale radio continuum surveys. While many recent studies attempted to classify source radio morphology from survey image data using deep learning algorithms (i.e., Convolutional Neural Networks), they concentrated on model robustness most time. It is unclear whether a model similarly makes predictions as radio astronomers did. In this work, we used Local Interpretable Model-agnostic Explanation (LIME), an state-of-the-art eXplainable Artificial Intelligence (XAI) technique to explain model prediction behaviour and thus examine the hypothesis in a proof-of-concept manner. In what follows, we describe how \textbf{LIME} generally works and early results about how it helped explain predictions of a radio galaxy classification model using this technique.

Explainable AI is an evolving area that deals with understanding the decision making of machine learning models so that these models are more transparent, accountable, and understandable for humans. In particular, post-hoc model-agnostic interpretable AI techniques explain the decisions of a black-box ML model for a single instance locally, without the knowledge of the intrinsic nature of the ML model. Despite their simplicity and capability in providing valuable insights, existing approaches fail to deliver consistent and reliable explanations. Moreover, in the context of black-box classifiers, existing approaches justify the predicted class, but these methods do not ensure that the explanation scores strongly differ as compared to those of another class. In this work we propose a novel post-hoc model agnostic XAI technique that provides contrastive explanations justifying the classification of a black box classifier along with a reasoning as to why another class was not predicted. Our method, which we refer to as CLIMAX which is short for Contrastive Label-aware Influence-based Model Agnostic XAI, is based on local classifiers . In order to ensure model fidelity of the explainer, we require the perturbations to be such that it leads to a class-balanced surrogate dataset. Towards this, we employ a label-aware surrogate data generation method based on random oversampling and Gaussian Mixture Model sampling. Further, we propose influence subsampling in order to retaining effective samples and hence ensure sample complexity. We show that we achieve better consistency as compared to baselines such as LIME, BayLIME, and SLIME. We also depict results on textual and image based datasets, where we generate contrastive explanations for any black-box classification model where one is able to only query the class probabilities for an instance of interest.

How to detect and mitigate deceptive AI systems is an open problem for the field of safe and trustworthy AI. We analyse two algorithms for mitigating deception: The first is based on the path-specific objectives framework where paths in the game that incentivise deception are removed. The second is based on shielding, i.e., monitoring for unsafe policies and replacing them with a safe reference policy. We construct two simple games and evaluate our algorithms empirically. We find that both methods ensure that our agent is not deceptive, however, shielding tends to achieve higher reward.