论文检索

Agathe Fernandes Machado, Ewen Gallic, Arthur Charpentier

Recently, optimal transport-based approaches have gained attention for deriving counterfactuals, e.g., to quantify algorithmic discrimination. However, in the general multivariate setting, these methods are often opaque and difficult to interpret. To address this, alternative methodologies have been proposed, using causal graphs combined with iterative quantile regressions or sequential transport to examine fairness at the individual level, often referred to as "counterfactual fairness." Despite these advancements, transporting categorical variables remains a significant challenge in practical applications with real datasets. In this paper, we propose a novel approach to address this issue. Our method involves (1) converting categorical variables into compositional data and (2) transporting these compositions within the probabilistic simplex of the Euclidean space. We demonstrate the applicability and effectiveness of this approach through an illustration on real-world data, and discuss limitations.

Amin Falah, Shibashis Guha, Ashutosh Trivedi

Reinforcement Learning (RL) is a sampling-based method for sequential decision-making, in which a learning agent iteratively converges toward an optimal policy by leveraging feedback from the environment in the form of scalar reward signals. While timing information is often abstracted in discrete-time domains, time-critical learning applications—such as queuing systems, population processes, and manufacturing systems—are naturally modeled as Continuous-Time Markov Decision Processes (CTMDPs). Since the seminal work of Bradtke and Duff, model-free RL for CTMDPs has become well-understood. However, in many practical applications, practitioners possess high-quality information about system rates derived from traditional queuing theory, which learning agents could potentially exploit to accelerate convergence. Despite this, classical RL algorithms for CTMDPs typically re-learn these parameters through sampling. In this work, we propose continuous-time reward machines (CTRMs), a novel framework that embeds reward functions and real-time state-action dynamics into a unified structure. CTRMs enable RL agents to effectively navigate dense-time environments while leveraging reward shaping and counterfactual experiences for accelerated learning. Our empirical results demonstrate CTRMs' ability to improve learning efficiency in time-critical environments.

Shuyang Dong, Shangtong Zhang, Lu Feng

Reinforcement Learning (RL) has shown great promise in domains like healthcare and robotics but often struggles with adoption due to its lack of interpretability. Counterfactual explanations, which address ``what if” scenarios, provide a promising avenue for understanding RL decisions but remain underexplored for continuous action spaces. We propose a novel approach for generating counterfactual explanations in continuous action RL by computing alternative action sequences that improve outcomes while minimizing deviations from the original sequence. Our approach leverages a distance metric for continuous actions and accounts for constraints such as adhering to predefined policies in specific states. Evaluations in two RL domains, Diabetes Control and Lunar Lander, demonstrate the effectiveness, efficiency, and generalization of our approach, enabling more interpretable and trustworthy RL applications.

Martin C. Cooper, Imane Bousdira, Clément Carbonnel

A classifier is considered interpretable if each of its decisions has an explanation which is small enough to be easily understood by a human user. A DNF can be seen as a binary classifier kappa over boolean domains. The size of an explanation of a positive decision taken by a DNF kappa is bounded by the size of the terms in kappa, since we can explain a positive decision by giving a term of kappa that evaluates to true. Since both positive and negative decisions must be explained, we consider that interpretable DNFs are those kappa for which both kappa and its complement can be expressed as DNFs composed of terms of bounded size. In this paper, we investigate the family of k-DNFs whose complements can also be expressed as k-DNFs. We compare two such families, namely depth-k decision trees and nested k-DNFs, a novel family of models. Experimental evidence indicates that nested k-DNFs are an interesting alternative to decision trees in terms of interpretability and accuracy.

Nasim Baharisangari, Yucheng Ruan, Chengcheng Zhao, Zhe Xu

Causal discovery in time-series datasets is critical for understanding complex systems, especially when the \textit{effectiveness} of causal relationships depends on both the \textit{duration} and \textit{magnitude} of the cause. We introduce a novel framework for causal discovery based on \textbf{Signal Temporal Logic (STL)}, enabling the extraction of interpretable causal diagrams (STL-CD) that explicitly capture these temporal dynamics. Our method first identifies statistically meaningful time intervals, then infers STL formulas that classify system behaviors, and finally employs transfer entropy to determine direct causal relationships among the formulas. This approach not only uncovers causal structure but also identifies the temporal persistence required for causal influence—an insight missed by existing methods. Experimental results on synthetic and real-world datasets demonstrate that our method achieves superior structural accuracy over state-of-the-art baselines, providing more informative and temporally precise causal models.

Yisong Wang, Xianglong Wang, Zhongtao Xie, Thomas Eiter

Explanation plays an important role in the decisions of both symbolic and neural network-based AI systems. Logic programs under answer set semantics (ASP) have been a typical declarative reasoning and problem-solving paradigm that has extensive applications in various AI domains. In this paper, we consider the issue of explanation for logic programs with abstract constraint atoms (c-atoms) under SPT-answer set semantics. Such c-atoms are general enough to capture complex constructors of logic programs, including aggregates, and the SPT-answer sets exclude circular justifications that other semantics have. We propose a minimal reduct for logic programs with c-atoms that yields a new semantic characterization of SPT-answer sets, and then introduce an extension of resolution for clauses with c-atoms. As we show, every atom in an SPT-answer set enjoys an extended resolution proof from the minimal reduct of its logic program. Finally, we present minimal sufficient subsets of logic programs (witnesses) to structure such an extended resolution proof for an atom in an SPT-answer set. Our results contribute to the justification of answer sets and provide a basis for explainability of ASP-based applications.

Christian Straßer, Badran Raddaoui, Said Jabbour

Inconsistency measures quantify the degree of conflict within a set of propositions. They can be broadly categorized into global measures, which assess the overall inconsistency of a set, and local measures, which evaluate the contribution of single formulas to the overall inconsistency. This paper investigates the relationship between these two classes of measures through the lens of marginal contributions and pooling mechanisms. We propose a systematic framework for deriving local inconsistency measures from global ones by employing notions of marginal contributions inspired by cooperative game theory, including Shapley and Banzhaf values. Conversely, we explore methods for constructing global inconsistency measures by aggregating local contributions using various pooling techniques. A key research question arises: which combinations of marginal contribution notions (maC) and pooling mechanisms (P) are compatible? Compatibility is defined such that, given a global measure I, applying (P) to the marginal contributions derived from I yields the same result as directly applying I, and vice versa. We analyze this compatibility condition and identify specific pairs of methods, (maC) and (P), that satisfy it across various inconsistency frameworks. Our findings provide a deeper understanding of the interplay between global and local inconsistency measures, providing a foundation for designing principled and interpretable inconsistency evaluation methods in logic-based systems.

Valeria Fionda, Antonio Ielo, Francesco Ricca

Recent advancements in AI have made LLMs valuable tools for automating the interpretation of textual descriptions of business processes and for converting formal process specifications into natural language. However, there are no practical methodologies or systematic assessments to ensure these automatic translations are faithful. This paper proposes a novel approach, based on an auxiliary bidirectional translation task, to assess LLMs performance quantitatively; also, it also empirically evaluates the performance of state-of-the-art LLMs for bidirectional translations between natural language and declarative formal process specifications. The results reveal substantial variability in performance among the LLMs, highlighting the importance of LLM selection and confirming the need for a robust method for assessing LLMs' outputs.

Gianvincenzo Alfano, Adam Gould, Francesco Leofante, Antonio Rago, Francesca Toni

Counterfactual explanations (CXs) are widely recognised as an essential technique for providing recourse recommendations for AI models. However, it is not obvious how to determine CXs in model multiplicity scenarios, where equally performing but different models can be obtained for the same task. In this paper, we propose novel qualitative and quantitative definitions of CXs based on explicit, nested quantification over (groups) of model decisions. We also study properties of these notions and identify decision problems of interest therefor. While our CXs are broadly applicable, in this paper we instantiate them within computational argumentation where model multiplicity naturally emerges, e.g. with incomplete and case-based argumentation frameworks. We then illustrate the suitability of our CXs for model multiplicity in legal and healthcare contexts, before analysing the complexity of the associated decision problems.

Shengbing Pei, Jiajun Ma, Zhao Lv, Chao Zhang, Jihong Guan

Abnormal brain functional network is an effective biomarker for brain disease diagnosis. Most existing methods focus on mining discriminative information from whole-brain connectivity patterns. However, multi-level collaboration is the foundation of efficient brain function, in addition to the whole-brain network, there are multiple sub-networks that can quickly integrate and process specific cognitive functions, forming the modular community structure of the brain. To address this gap, we propose a novel method, community-aware graph Transformer (CAGT), that integrates the community information of sub-networks and the topological information of brain graph into the Transformer architecture for better brain disorder identification. CAGT enhances information exchange within and between functional communities through dual-scale feature fusion, capturing interactive information across various scales. Additionally, it incorporates prior knowledge to design brain region position encoding and guide the self-attention, thereby enhancing the spatial awareness of the Transformer and aligning it with the brain's natural information transfer process. Experimental results indicate that our proposed method significantly improves performance on both large and small datasets, and can reliably capture the interactions between sub-networks, demonstrating its generalization and interpretability.

Yuchen Wang, Dongpeng Hou, Chao Gao, Xianghua Li

Multivariate Temporal Point Processes (MTPPs) play an important role in diverse domains such as social networks and finance for predicting event sequence data. In recent years, MTPPs based on Ordinary Differential Equations (ODEs) and Stochastic Differential Equations (SDEs) have demonstrated their strong modeling capabilities. However, these models have yet to thoroughly consider the underlying relationships among different event types to enhance their modeling capacity. Therefore, this paper introduces a method that uses neural SDEs with a jump process guided by the latent graph. Firstly, our proposed method employs multi-dimensional SDEs to capture the dynamics of the intensity function for each event type. Subsequently, a latent graph structure is integrated into the jump process without any encoder, aiming to enhance the modeling and predictive capabilities for MTPPs. Theoretical analysis guarantees the existence and uniqueness of the solution for our proposed method. The experiments conducted on multiple real-world datasets show that our approaches demonstrate significant competitiveness when compared to state-of-the-art neural point processes. Meanwhile, the trainable parameters of the latent graph also improve the model interpretability without any prior knowledge. Our code is available at https://github.com/cgao-comp/LNJSDE.

Xudong Wang, Ziheng Sun, Chris Ding, Jicong Fan

Explainable artificial intelligence (XAI) is an important area in the AI community, and interpretability is crucial for building robust and trustworthy AI models. While previous work has explored model-level and instance-level explainable graph learning, there has been limited investigation into explainable graph representation learning. In this paper, we focus on representation-level explainable graph learning and ask a fundamental question: What specific information about a graph is captured in graph representations? Our approach is inspired by graph kernels, which evaluate graph similarities by counting substructures within specific graph patterns. Although the pattern counting vector can serve as an explainable representation, it has limitations such as ignoring node features and being high-dimensional. To address these limitations, we introduce a framework (PXGL-GNN) for learning and explaining graph representations through graph pattern analysis. We start by sampling graph substructures of various patterns. Then, we learn the representations of these patterns and combine them using a weighted sum, where the weights indicate the importance of each graph pattern's contribution. We also provide theoretical analyses of our methods, including robustness and generalization. In our experiments, we show how to learn and explain graph representations for real-world data using pattern analysis. Additionally, we compare our method against multiple baselines in both supervised and unsupervised learning tasks to demonstrate its effectiveness.

Lingxiao Wang, Lei Shi, Feifei Kou, Ligu Zhu, Chen Ma, Pengfei Zhang, Mingying Xu, Zeyu Li

Large Language Models (LLMs) and Retrieval-Augmented Generation (RAG) have demonstrated significant potential in automated fact-checking. However, existing methods face limitations in insufficient evidence utilization and lack of explicit verification criteria. Specifically, these approaches aggregate evidence for collective reasoning without independently analyzing each piece, hindering their ability to leverage the available information thoroughly. Additionally, they rely on simple prompts or few-shot learning for verification, which makes truthfulness judgments less reliable, especially for complex claims. To address these limitations, we propose a novel method to enhance evidence utilization and introduce explicit verification criteria, named EVICheck. Our approach independently reasons each evidence piece and synthesizes the results to enable more thorough exploration and enhance interpretability. Additionally, by incorporating fine-grained truthfulness criteria, we make the model's verification process more structured and reliable, especially when handling complex claims. Experimental results on the public RAWFC dataset demonstrate that EVICheck achieves state-of-the-art performance across all evaluation metrics. Our method demonstrates strong potential in fake news verification, significantly improving the accuracy.

Hao Wang, Jindong Han, Wei Fan, Leilei Sun, Hao Liu

Spatio-temporal forecasting is pivotal in numerous real-world applications, including transportation planning, energy management, and climate monitoring. In this work, we aim to harness the reasoning and generalization abilities of Pre-trained Language Models (PLMs) for more effective spatio-temporal forecasting, particularly in data-scarce scenarios. However, recent studies uncover that PLMs, which are primarily trained on textual data, often falter when tasked with modeling the intricate correlations in numerical time series, thereby limiting their effectiveness in comprehending spatio-temporal data. To bridge the gap, we propose RePST, a semantic-oriented PLM reprogramming framework tailored for spatio-temporal forecasting. Specifically, we first propose a semantic-oriented decomposer that adaptively disentangles spatially correlated time series into interpretable sub-components, which facilitates PLM to understand sophisticated spatio-temporal dynamics via a divide-and-conquer strategy. Moreover, we propose a selective discrete reprogramming scheme, which introduces an expanded spatio-temporal vocabulary space to project spatio-temporal series into discrete representations. This scheme minimizes the information loss during reprogramming and enriches the representations derived by PLMs. Extensive experiments on real-world datasets show that the proposed RePST outperforms twelve state-of-the-art baseline methods, particularly in data-scarce scenarios, highlighting the effectiveness and superior generalization capabilities of PLMs for spatio-temporal forecasting. Codes and Appendix can be found at https://github.com/usail-hkust/REPST.

Lijia Ma, Haoyang Fu, Zhijie Cao, Xiongnan Jin, Qiuzhen Lin, Jianqiang Li

Link Sign Prediction (LSP) in signed networks is a critical task with applications in recommendation systems, community detection, and social network analysis. Existing methods primarily rely on graph neural networks to exploit structural information, often neglecting the valuable insights from edge-level textual data. Furthermore, utilizing large language models (LLMs) for LSP faces challenges in reliability and interpreting graph structures. To address these issues, we propose a novel STLSP framework that integrates signed networks' \underline{S}tructural and \underline{T}extual information with LLMs for the \underline{LSP} task. STLSP leverages structural balance theory to generate node embeddings that capture positive and negative relationships. These embeddings are transformed into natural language representations through clustering techniques, allowing LLMs to utilize the structural context fully. By integrating these representations with edge text, STLSP improves the accuracy and reliability of the LSP task. Extensive experiments conducted on five real-world datasets demonstrate that STLSP outperformed state-of-the-art baselines, achieving an 8.7% improvement in terms of accuracy. Moreover, STLSP shows robust performance across various LLMs, making it adaptable to different computational environments. The code and data are publically available at https://github.com/sss483/STLSP.

Yezi Liu, Jiaxuan Xie, Yanning Shen

Dynamic graph neural networks (dynamic GNNs) have demonstrated remarkable effectiveness in analyzing time-varying graph-structured data. However, their black-box nature often hinders users from understanding their predictions, which can limit their applications. In recent years, there has been a surge in research aimed at explaining GNNs, but most studies have focused on static graphs, leaving the explanation of dynamic GNNs relatively unexplored. Explaining dynamic GNNs presents a unique challenge due to their complex spatial and temporal structures. As a result, existing approaches designed for explaining static graphs are not directly applicable to dynamic graphs because they ignore temporal dependencies among graph snapshots. To address this issue, we propose DGExplainer, which offers a reliable explanation of dynamic GNN predictions. DGExplainer utilizes the relevance back-propagation technique both time-wise and layer-wise. Specifically, it incorporates temporal information by computing the relevance of node representations along the inverse of the time evolution. Additionally, for each time step, it calculates layer-wise relevance from a graph-based module by redistributing the relevance of node representations along the back-propagation path. Quantitative and qualitative experimental results on six real-world datasets demonstrate the effectiveness of DGExplainer in identifying important nodes for link prediction and node regression in dynamic GNNs.

Joao Marques-Silva, Jairo A. Lefebre-Lobaina, Maria Vanina Martinez

Explainable artificial intelligence (XAI) is at the core of trustworthy AI. The best-known methods of XAI are sub-symbolic. Unfortunately, these methods do not give guarantees of rigor. Logic-based XAI addresses the lack of rigor of sub-symbolic methods, but in turn it exhibits some drawbacks. These include scalability, explanation size, but also the need to access the details of the machine learning model. Furthermore, access to the details of an ML model may reveal sensitive information. This paper builds on recent work on symbolic model-agnostic XAI, which is based on explaining samples of behavior of a blackbox ML model, and proposes efficient algorithms for the computation of explanations. The experiments confirm the scalability of the novel algorithms.

Florent Avellaneda

Decision trees are widely used in machine learning for their interpretability and effectiveness in classification tasks. Traditional axis-parallel decision trees partition data using single-feature thresholds at each node, but they often struggle to represent complex, non-axis-aligned decision boundaries efficiently. This limitation can result in unnecessarily large and less interpretable trees. Oblique decision trees address this limitation by using linear combinations of features at each node, allowing a more natural representation of complex decision boundaries while maintaining interpretability through sparse linear combinations. However, learning optimal oblique decision trees poses a significant computational challenge, as existing methods predominantly rely on suboptimal greedy heuristics. In this paper, we propose a novel approach to learning globally optimal oblique decision trees by reformulating the problem as a (Max)SAT instance. By leveraging state-of-the-art (Max)SAT solvers, our method efficiently explores the solution space to identify optimal trees. Experiments on benchmark datasets demonstrate that our approach generates optimal oblique decision trees within reasonable computational time for small to medium-sized datasets.

Binjia Zhou, Hengrui Lou, Lizhe Chen, Haoyuan Li, Dawei Luo, Shuai Chen, Jie Lei, Zunlei Feng, Yijun Bei

With the swift progression of image generation technology, the widespread emergence of facial deepfakes poses significant challenges to the field of security, thus amplifying the urgent need for effective deepfake detection. Existing techniques for face forgery detection can broadly be categorized into two primary groups: visual-based methods and multimodal approaches. The former often lacks clear explanations for forgery details, while the latter, which merges visual and linguistic modalities, is more prone to the issue of hallucinations.To address these shortcomings, we introduce a visual detail enhanced self-correction framework, designated CorrDetail, for interpretable face forgery detection. CorrDetail is meticulously designed to rectify authentic forgery details when provided with error-guided questioning, with the aim of fostering the ability to uncover forgery details rather than yielding hallucinated responses. Additionally, to bolster the reliability of its findings, a visual fine-grained detail enhancement module is incorporated, supplying CorrDetail with more precise visual forgery details. Ultimately, a fusion decision strategy is devised to further augment the model's discriminative capacity in handling extreme samples, through the integration of visual information compensation and model bias reduction. Experimental results demonstrate that CorrDetail not only achieves state-of-the-art performance compared to the latest methodologies but also excels in accurately identifying forged details, all while exhibiting robust generalization capabilities.

Zhemin Zhang, Xun Gong

Foundation models, such as CNNs and ViTs, have powered the development of image representation learning. However, general guidance to model architecture design is still missing. Inspired by the connection between image representation learning and heat conduction, we model images by the heat conduction equation, where the essential idea is to conceptualize image features as temperatures and model their information interaction as the diffusion of thermal energy. Based on this idea, we find that many modern model architectures, such as residual structures, SE block, and feed-forward networks, can be interpreted from the perspective of the heat conduction equation. Therefore, we leverage the heat equation to design new and more interpretable models. As an example, we propose the Heat Conduction Layer and the Refinement Approximation Layer inspired by solving the heat conduction equation using Finite Difference Method and Fourier series, respectively. The main goal of this paper is to integrate the overall architectural design of neural networks into the theoretical framework of heat conduction. Nevertheless, our Heat Conduction Network (HcNet) still shows competitive performance, e.g., HcNet-T achieves 83.0% top-1 accuracy on ImageNet-1K while only requiring 28M parameters and 4.1G MACs. The code is publicly available at: https://github.com/ZheminZhang1/HcNet.