Contrastive Explanation Research Articles

CE-MRS: Contrastive Explanations for Multi-Robot Systems

CE-MRS: Contrastive Explanations for Multi-Robot Systems

A Novel Integration of Data-Driven Rule Generation and Computational Argumentation for Enhanced Explainable AI

Explainable Artificial Intelligence (XAI) is a research area that clarifies AI decision-making processes to build user trust and promote responsible AI. Hence, a key scientific challenge in XAI is the development of methods that generate transparent and interpretable explanations while maintaining scalability and effectiveness in complex scenarios. Rule-based methods in XAI generate rules that can potentially explain AI inferences, yet they can also become convoluted in large scenarios, hindering their readability and scalability. Moreover, they often lack contrastive explanations, leaving users uncertain why specific predictions are preferred. To address this scientific problem, we explore the integration of computational argumentation—a sub-field of AI that models reasoning processes through defeasibility—into rule-based XAI systems. Computational argumentation enables arguments modelled from rules to be retracted based on new evidence. This makes it a promising approach to enhancing rule-based methods for creating more explainable AI systems. Nonetheless, research on their integration remains limited despite the appealing properties of rule-based systems and computational argumentation. Therefore, this study also addresses the applied challenge of implementing such an integration within practical AI tools. The study employs the Logic Learning Machine (LLM), a specific rule-extraction technique, and presents a modular design that integrates input rules into a structured argumentation framework using state-of-the-art computational argumentation methods. Experiments conducted on binary classification problems using various datasets from the UCI Machine Learning Repository demonstrate the effectiveness of this integration. The LLM technique excelled in producing a manageable number of if-then rules with a small number of premises while maintaining high inferential capacity for all datasets. In turn, argument-based models achieved comparable results to those derived directly from if-then rules, leveraging a concise set of rules and excelling in explainability. In summary, this paper introduces a novel approach for efficiently and automatically generating arguments and their interactions from data, addressing both scientific and applied challenges in advancing the application and deployment of argumentation systems in XAI.

Evaluative Item-Contrastive Explanations in Rankings

The remarkable success of Artificial Intelligence in advancing automated decision-making is evident both in academia and industry. Within the plethora of applications, ranking systems hold significant importance in various domains. This paper advocates for the application of a specific form of Explainable AI—namely, contrastive explanations—as particularly well-suited for addressing ranking problems. This approach is especially potent when combined with an Evaluative AI methodology, which conscientiously evaluates both positive and negative aspects influencing a potential ranking. Therefore, the present work introduces Evaluative Item-Contrastive Explanations tailored for ranking systems and illustrates its application and characteristics through an experiment conducted on publicly available data.

Contrastive Explanations of Centralized Multi-agent Optimization Solutions

Contrastive Explanations of Centralized Multi-agent Optimization Solutions

CoRTEx: contrastive learning for representing terms via explanations with applications on constructing biomedical knowledge graphs.

Biomedical Knowledge Graphs play a pivotal role in various biomedical research domains. Concurrently, term clustering emerges as a crucial step in constructing these knowledge graphs, aiming to identify synonymous terms. Due to a lack of knowledge, previous contrastive learning models trained with Unified Medical Language System (UMLS) synonyms struggle at clustering difficult terms and do not generalize well beyond UMLS terms. In this work, we leverage the world knowledge from large language models (LLMs) and propose Contrastive Learning for Representing Terms via Explanations (CoRTEx) to enhance term representation and significantly improves term clustering. The model training involves generating explanations for a cleaned subset of UMLS terms using ChatGPT. We employ contrastive learning, considering term and explanation embeddings simultaneously, and progressively introduce hard negative samples. Additionally, a ChatGPT-assisted BIRCH algorithm is designed for efficient clustering of a new ontology. We established a clustering test set and a hard negative test set, where our model consistently achieves the highest F1 score. With CoRTEx embeddings and the modified BIRCH algorithm, we grouped 35 580 932 terms from the Biomedical Informatics Ontology System (BIOS) into 22 104 559 clusters with O(N) queries to ChatGPT. Case studies highlight the model's efficacy in handling challenging samples, aided by information from explanations. By aligning terms to their explanations, CoRTEx demonstrates superior accuracy over benchmark models and robustness beyond its training set, and it is suitable for clustering terms for large-scale biomedical ontologies.

Method in relational-explanatory geography

Human geography’s onto-epistemological expansion in recent decades has not been well matched by methodological development. Learning from the methodological pluralism in four relational approaches, this paper reflects on key issues in relational-explanatory theorizing before introducing a social science method of process tracing in relation to a comparative methodology. I argue that contrastive explanations can be developed through deploying comparable methodological practices as different ‘moments’ of a research process in appropriate evidential contexts. This research process-based comparative methodology can better trace actors and their relational networks through in situ research and generate explanatory theoretical insights into the complexity of socio-spatial life.

‘Why Didn’t You Allocate This Task to Them?’ Negotiation-Aware Task Allocation and Contrastive Explanation Generation

In this work, we design an Artificially Intelligent Task Allocator (AITA) that proposes a task allocation for a team of humans. A key property of this allocation is that when an agent with imperfect knowledge (about their teammate's costs and/or the team's performance metric) contests the allocation with a counterfactual, a contrastive explanation can always be provided to showcase why the proposed allocation is better than the proposed counterfactual. For this, we consider a negotiation process that produces a negotiation-aware task allocation and, when contested, leverages a negotiation tree to provide a contrastive explanation. With human subject studies, we show that the proposed allocation indeed appears fair to a majority of participants and, when not, the explanations generated are judged as convincing and easy to comprehend.

CrystalBox: Future-Based Explanations for Input-Driven Deep RL Systems

We present CrystalBox, a novel, model-agnostic, posthoc explainability framework for Deep Reinforcement Learning (DRL) controllers in the large family of input-driven environments which includes computer systems. We combine the natural decomposability of reward functions in input-driven environments with the explanatory power of decomposed returns. We propose an efficient algorithm to generate future-based explanations across both discrete and continuous control environments. Using applications such as adaptive bitrate streaming and congestion control, we demonstrate CrystalBox's capability to generate high-fidelity explanations. We further illustrate its higher utility across three practical use cases: contrastive explanations, network observability, and guided reward design, as opposed to prior explainability techniques that identify salient features.

Explainability of Speech Recognition Transformers via Gradient-Based Attention Visualization

In vision Transformers, attention visualization methods are used to generate heatmaps highlighting the class-corresponding areas in input images, which offers explanations on how the models make predictions. However, it is not so applicable for explaining automatic speech recognition (ASR) Transformers. An ASR Transformer makes a particular prediction for every input token to form a sentence, but a vision Transformer only makes an overall classification for the input data. Therefore, traditional attention visualization methods may fail in ASR Transformers. In this work, we propose a novel attention visualization method in ASR Transformers and try to explain which frames of the audio result in the output text. Inspired by the model explainability, we also explore ways of improving the effectiveness of the ASR model. Comparing with other Transformer attention visualization methods, our method is more efficient and intuitively understandable, which unravels the attention calculation from information flow of Transformer attention modules. In addition, we demonstrate the utilization of visualization result in three ways: (1) We visualize attention with respect to connectionist temporal classification (CTC) loss to train an ASR model with adversarial attention erasing regularization, which effectively decreases the word error rate (WER) of the model and improves its generalization capability. (2) We visualize the attention on some specific words, interpreting the model by effectively demonstrating the semantic and grammar relationships between these words. (3) Similarly, we analyze how the model manage to distinguish homophones, using contrastive explanation with respect to homophones.

VOICE: Variance of Induced Contrastive Explanations to quantify Uncertainty in Neural Network Interpretability

VOICE: Variance of Induced Contrastive Explanations to quantify Uncertainty in Neural Network Interpretability

Model Agnostic Contrastive Explanations for Classification Models

Model Agnostic Contrastive Explanations for Classification Models

Generation of Molecular Counterfactuals for Explainable Machine Learning Based on Core-Substituent Recombination.

The use of black box machine learning models whose decisions cannot be understood limits the acceptance of predictions in interdisciplinary research and camouflages artificial learning characteristics leading to predictions for other than anticipated reasons. Consequently, there is increasing interest in explainable artificial intelligence to rationalize predictions and uncover potential pitfalls. Among others, relevant approaches include feature attribution methods to identify molecular structures determining predictions and counterfactuals (CFs) or contrastive explanations. CFs are defined as variants of test instances with minimal modifications leading to opposing predictions. In medicinal chemistry, CFs have thus far only been little investigated although they are particularly intuitive from a chemical perspective. We introduce a new methodology for the systematic generation of CFs that is centered on well-defined structural analogues of test compounds. The approach is transparent, computationally straightforward, and shown to provide a wealth of CFs for test sets. The method is made freely available.

Enhancing anomaly detection accuracy and interpretability in low-quality and class imbalanced data: A comprehensive approach

With the rapid advancements in smart meters, energy internet, and high-performance computing technologies, deep learning methods are increasingly used for detecting anomalies in building electricity consumption. However, these models grapple with challenges like outliers, missing data, and imbalanced categories. Moreover, these models are often considered as black boxes due to their lack of interpretability. This limitation impedes the practical deployment of deep learning methods. Focusing on these issues, this study presents an anomaly detection method aimed at improving anomaly detection accuracy and interpretability in low-quality and class imbalanced data. Initially, our method established data preprocessing models to handle with outliers and missing data. Subsequently, an unbalance elimination model, named GRU-MACGANS, was designed to augment the minority category data. The SA-CNN-BiGRU-FCN model, incorporating a knowledge transfer strategy, was implemented to identify anomalous behavior. Finally, the results of SA-CNN-BiGRU-FCN model were interpreted based on the contrastive explanations method. Our case study of a low-voltage distribution system demonstrates the effectiveness of our approach. Data preprocessing models and GRU-MACGANS enhance data quality and alleviate category imbalance issues in anomaly detection. The interpretable analysis provides insights into the ability of detection models to identify representative features of different electricity consumption behaviors for effective anomaly detection.

Scaffolding the human partner by contrastive guidance in an explanatory human-robot dialogue.

Explanation has been identified as an important capability for AI-based systems, but research on systematic strategies for achieving understanding in interaction with such systems is still sparse. Negation is a linguistic strategy that is often used in explanations. It creates a contrast space between the affirmed and the negated item that enriches explaining processes with additional contextual information. While negation in human speech has been shown to lead to higher processing costs and worse task performance in terms of recall or action execution when used in isolation, it can decrease processing costs when used in context. So far, it has not been considered as a guiding strategy for explanations in human-robot interaction. We conducted an empirical study to investigate the use of negation as a guiding strategy in explanatory human-robot dialogue, in which a virtual robot explains tasks and possible actions to a human explainee to solve them in terms of gestures on a touchscreen. Our results show that negation vs. affirmation 1) increases processing costs measured as reaction time and 2) increases several aspects of task performance. While there was no significant effect of negation on the number of initially correctly executed gestures, we found a significantly lower number of attempts-measured as breaks in the finger movement data before the correct gesture was carried out-when being instructed through a negation. We further found that the gestures significantly resembled the presented prototype gesture more following an instruction with a negation as opposed to an affirmation. Also, the participants rated the benefit of contrastive vs. affirmative explanations significantly higher. Repeating the instructions decreased the effects of negation, yielding similar processing costs and task performance measures for negation and affirmation after several iterations. We discuss our results with respect to possible effects of negation on linguistic processing of explanations and limitations of our study.

Model-contrastive explanations through symbolic reasoning

Explaining how two machine learning classification models differ in their behaviour is gaining significance in eXplainable AI, given the increasing diffusion of learning-based decision support systems. Human decision-makers deal with more than one machine learning model in several practical situations. Consequently, the importance of understanding how two machine learning models work beyond their prediction performances is key to understanding their behaviour, differences, and likeness.Some attempts have been made to address these problems, for instance, by explaining text classifiers in a time-contrastive fashion. In this paper, we present MERLIN, a novel eXplainable AI approach that provides contrastive explanations of two machine learning models, introducing the concept of model-contrastive explanations. We propose an encoding that allows MERLIN to work with both text and tabular data and with mixed continuous and discrete features. To show the effectiveness of our approach, we evaluate it on an extensive set of benchmark datasets. MERLIN is also implemented as a python-pip package.

Exploring English-Spanish contrastive grammar explanations in English for Academic Purposes reading comprehension materials at university level

A topic of debate in English for Academic Purposes reading comprehension courses in Spanish speaking universities is the roleof grammar in in-house materials. This study explored and analysed the way in which these materials deal with contrastive grammar, in order to promote reflection and discussion on the subject of English-Spanish contrastive grammar explanations. Results showed that there are inaccuracies in the metalanguage used, in the Spanish-English contrastive analysis of some morphosyntactic structures and in the choices of instances of direct translation from English into Spanish. It is concluded that Spanish teachers of English academic reading comprehension should be provided with reference materials that allow them to describe in a well-founded manner the differences and similarities between the linguistic resources used in these two languages. Un tema de debate en los cursos de comprensión lectora de Inglés con Fines Académicos en las universidades de habla hispana es el papel de la gramática en los materiales internos. En este estudio se explora y analiza la forma en que estos materiales abordan la gramática contrastiva, con el fin de promover la reflexión y el debate sobre el tema de las explicaciones gramaticales contrastivas inglés-español. Los resultados mostraron que existen imprecisiones en el metalenguaje utilizado, en el análisis contrastivo español-inglés de algunas estructuras morfosintácticas y en la elección de instancias de traducción directa del inglés al español. Se concluye que los profesores españoles de comprensión lectora de Inglés con Fines Académicos deben contar con materiales de referencia que les permitan describir de manera fundamentada las diferencias y similitudes entre los recursos lingüísticos utilizados en estas dos lenguas.

Human performance consequences of normative and contrastive explanations: An experiment in machine learning for reliability maintenance

Decision aids based on artificial intelligence and machine learning can benefit human decisions and system performance, but can also provide incorrect advice, and invite operators to inappropriately rely on automation. This paper examined the extent to which example-based explanations could improve reliance on a decision aid that is based on machine learning. Participants engaged in a preventive maintenance task by providing their diagnosis of the conditions of three components of a hydraulic system. A decision aid based on machine learning provided advice but was not always reliable. Three explanation displays (baseline, normative, normative plus contrastive) were manipulated within-participants. With the normative explanation display, we found improvements in participants' decision time and subjective workload. With the addition of contrastive explanations, we found improvements in participants' hit rate and sensitivity in discriminating between correct and incorrect ML advice. Implications for the design of explainable interfaces to support human-AI interaction in data intensive environments are discussed.

Accurate clinical toxicity prediction using multi-task deep neural nets and contrastive molecular explanations

Explainable machine learning for molecular toxicity prediction is a promising approach for efficient drug development and chemical safety. A predictive ML model of toxicity can reduce experimental cost and time while mitigating ethical concerns by significantly reducing animal and clinical testing. Herein, we use a deep learning framework for simultaneously modeling in vitro, in vivo, and clinical toxicity data. Two different molecular input representations are used; Morgan fingerprints and pre-trained SMILES embeddings. A multi-task deep learning model accurately predicts toxicity for all endpoints, including clinical, as indicated by the area under the Receiver Operator Characteristic curve and balanced accuracy. In particular, pre-trained molecular SMILES embeddings as input to the multi-task model improved clinical toxicity predictions compared to existing models in MoleculeNet benchmark. Additionally, our multitask approach is comprehensive in the sense that it is comparable to state-of-the-art approaches for specific endpoints in in vitro, in vivo and clinical platforms. Through both the multi-task model and transfer learning, we were able to indicate the minimal need of in vivo data for clinical toxicity predictions. To provide confidence and explain the model’s predictions, we adapt a post-hoc contrastive explanation method that returns pertinent positive and negative features, which correspond well to known mutagenic and reactive toxicophores, such as unsubstituted bonded heteroatoms, aromatic amines, and Michael receptors. Furthermore, toxicophore recovery by pertinent feature analysis captures more of the in vitro (53%) and in vivo (56%), rather than of the clinical (8%), endpoints, and indeed uncovers a preference in known toxicophore data towards in vitro and in vivo experimental data. To our knowledge, this is the first contrastive explanation, using both present and absent substructures, for predictions of clinical and in vivo molecular toxicity.

XAI USING FORMAL CONCEPT LATTICE FOR IMAGE DATA

A Formal concept lattice can be used to generate explanations from a black box model. This novel approach has been applied and proven for tabular data. It has also been compared to popular techniques in XAI. In this work, we apply the approach to image data. Image data, in general, comprises large dimensions and hence poses a challenge to build a formal concept lattice from such large data. We break the image into parts and build multiple sub-lattices. Using the combination of sub-lattice explanations, we generate the complete explanation for the entire image. We present our work beginning with a simple synthetic dataset for providing an intuitive idea of explanations and its credibility. This is followed by explanations of a model built on the popular MNIST dataset proving consistency of explanations on a real dataset. Text explanations from the lattice are converted to images for ease of visual understanding. We compare our work with DeepLIFT by viewing image masks obtained through contrastive explanation for specific digits from the MNIST dataset. This work proves the feasibility of using formal concept lattices for image data.

A unified logical framework for explanations in classifier systems

AbstractRecent years have witnessed a renewed interest in the explanation of classifier systems in the field of explainable AI (XAI). The standard approach is based on propositional logic. We present a modal language which supports reasoning about binary input classifiers and their properties. We study a family of classifier models, axiomatize it as two proof systems regarding the cardinality of the language and show completeness of our axiomatics. Moreover, we show that the satisfiability checking problem for our modal language is NEXPTIME-complete in the infinite-variable case, while it becomes polynomial in the finite-variable case. We moreover identify an interesting NP fragment of our language in the infinite-variable case. We leverage the language to formalize counterfactual conditional as well as a variety of notions of explanation including abductive, contrastive and counterfactual explanations and biases. Finally, we present two extensions of our language: a dynamic extension by the notion of assignment enabling classifier change and an epistemic extension in which the classifier’s uncertainty about the actual input can be represented.