Identifying the causal pathways of unfairness is a critical objective for improving policy design and algorithmic decision-making. Prior work in causal fairness analysis often requires knowledge of the causal graph, hindering practical applications in complex or low-knowledge domains. Moreover, global discovery methods that learn causal structure from data can display unstable performance on finite samples, preventing robust fairness conclusions. To mitigate these challenges, we introduce local discovery for direct discrimination (LD3): a method that uncovers structural evidence of direct unfairness by identifying the causal parents of an outcome variable. LD3 performs a linear number of conditional independence tests relative to variable set size, and allows for latent confounding under the sufficient condition that all parents of the outcome are observed. We show that LD3 returns a valid adjustment set (VAS) under a new graphical criterion for the weighted controlled direct effect, a qualitative indicator of direct discrimination. LD3 limits unnecessary adjustment, providing interpretable VAS for assessing unfairness. We use LD3 to analyze causal fairness in two complex decision systems: criminal recidivism prediction and liver transplant allocation. LD3 was more time-efficient and returned more plausible results on real-world data than baselines, which took 46× to 5870× longer to execute.
论文检索
输入标题、作者或关键词，从 7,537 篇学术成果中精准定位
Cross-Domain Few-Shot Learning (CDFSL) requires the model to transfer knowledge from the data-abundant source domain to data-scarce target domains for fast adaptation, where the large domain gap makes CDFSL a challenging problem. Masked Autoencoder (MAE) excels in effectively using unlabeled data and learning image’s global structures, enhancing model generalization and robustness. However, in the CDFSL task with significant domain shifts, we find MAE even shows lower performance than the baseline supervised models. In this paper, we first delve into this phenomenon for an interpretation. We find that MAE tends to focus on low-level domain information during reconstructing pixels while changing the reconstruction target to token features could mitigate this problem. However, not all features are beneficial, as we then find reconstructing high-level features can hardly improve the model’s transferability, indicating a trade-off between filtering domain information and preserving the image’s global structure. In all, the reconstruction target matters for the CDFSL task. Based on the above findings and interpretations, we further propose Domain-Agnostic Masked Image Modeling (DAMIM) for the CDFSL task. DAMIM includes an Aggregated Feature Reconstruction module to automatically aggregate features for reconstruction, with balanced learning of domain-agnostic information and images’ global structure, and a Lightweight Decoder module to further benefit the encoder’s generalizability. Experiments on four CDFSL datasets demonstrate that our method achieves state-of-the-art performance.
Multivariate time series anomaly detection has numerous real-world applications and is being extensively studied. Modeling pairwise correlations between variables is crucial. Existing methods employ learnable graph structures and graph neural networks to explicitly model the spatial dependencies between variables. However, these methods are primarily based on prediction or reconstruction tasks, which can only learn similarity relationships between sequence embeddings and lack interpretability in how graph structures affect time series evolution. In this paper, we designed a framework that models spatial dependencies using interpretable causal relationships and detects anomalies through changes in causal patterns. Specifically, we propose a method to dynamically discover Granger causality using gradients in nonlinear deep predictors and employ a simple sparsification strategy to obtain a Granger causality graph, detecting anomalies from a causal perspective. Experiments on real-world datasets demonstrate that the proposed model achieves more accurate anomaly detection compared to baseline methods.
Differential equations are widely used to describe complex dynamical systems with evolving parameters in nature and engineering. Effectively learning a family of maps from the parameter function to the system dynamics is of great significance. In this study, we propose a novel learning framework of symbolic continuous-depth neural networks, termed Symbolic Neural Ordinary Differential Equations (SNODEs), to effectively and accurately learn the underlying dynamics of complex systems. Specifically, our learning framework comprises three stages: initially, pre-training a predefined symbolic neural network via a gradient flow matching strategy; subsequently, fine-tuning this network using Neural ODEs; and finally, constructing a general neural network to capture residuals. In this process, we apply the SNODEs framework to partial differential equation systems through Fourier analysis, achieving resolution-invariant modeling. Moreover, this framework integrates the strengths of symbolism and connectionism, boasting a universal approximation theorem while significantly enhancing interpretability and extrapolation capabilities relative to state-of-the-art baseline methods. We demonstrate this through experiments on several representative complex systems. Therefore, our framework can be further applied to a wide range of scientific problems, such as system bifurcation and control, reconstruction and forecasting, as well as the discovery of new equations.
Unravelling Causal Genetic Biomarkers of Alzheimer’s Disease via Neuron to Gene-token Backtracking in Neural Architecture: A Groundbreaking Reverse-Gene-Finder Approach
PDF ↗Alzheimer’s Disease (AD) affects over 55 million people globally, yet the key genetic contributors remain poorly understood. Leveraging recent advancements in genomic foundation models, we present the innovative Reverse-Gene-Finder technology, a ground-breaking neuron-to-gene-token backtracking approach in a neural network architecture to elucidate the novel causal genetic biomarkers driving AD onset. Reverse-Gene-Finder comprises three key innovations. Firstly, we exploit the observation that genes with the highest probability of causing AD, defined as the most causal genes (MCGs), must have the highest probability of activating those neurons with the highest probability of causing AD, defined as the most causal neurons (MCNs). Secondly, we utilize a gene token representation at the input layer to allow each gene (known or novel to AD) to be represented as a discrete and unique entity in the input space. Lastly, in contrast to the existing neural network architectures, which track neuron activations from the input layer to the output layer in a feed-forward manner, we develop an innovative backtracking method to track backwards from the MCNs to the input layer, identifying the Most Causal Tokens (MCTs) and the corresponding MCGs. Reverse-Gene-Finder is highly interpretable, generalizable, and adaptable, providing a promising avenue for application in other disease scenarios.
In recent Open-set Recognition (OSR) community, a prevailing belief is that enhancing the discriminative boundaries of closed-set classes can improve the robustness of Deep Neural Networks (DNNs) against open data during testing. Typical studies validate this *implicitly* by empirical evidence, without a formalized understanding of *how DNNs help the closed-set features obtain more discriminative boundaries?* For this, we provide an answer from the Neural Collapse (NC) perspective: DNNs align the closed-set with a *Simplex Equiangular Tight Frame* (ETF) structure that has geometric and mathematical interpretability. Regrettably, although NC naturally occurs in DNNs, we discover that typical studies cannot guarantee the features being learned to strictly align with the ETF. Thus, we introduce a novel concept, Fixed ETF Template (FiT), which holds an ideal structure associated with closed-set classes. To force class means and classifier vectors to align with FiT, we further design a Dual ETF (DEF) loss involving two components. Specifically, *F*-DEF loss is designed to align class means with FiT strictly, yielding optimal inter-class separability. Meanwhile, we extend a dual form to classifier vectors, termed *C*-DEF loss, which guides class means and classifier vectors to satisfy self-duality. Our theoretical analysis proves the validity of the proposed approach, and extensive experiments demonstrate that DEF achieves comparable or superior results with reduced computational resources on standard OSR benchmarks.
HYBOOD: A Hybrid Generative Model for Out-of-Distribution Detection with Corruption Estimation
PDF ↗We propose HYBOOD, a hybrid out-of-distribution model based on normalizing flow followed by a simple linear classification model. In real-world settings, it is known that data corruption has a strong influence on model degradation; for example image quality like noise, blur and image geometry like translation, scaling, rotation. MNIST-C, CIFAR10-C are the general synthesized datasets to measure model performance and corruption difficulty in terms of covariate and semantic shifts. HYBOOD shows that the separability between in-distribution, covariate shift, and semantic shift can be represented by distribution distance and log-scale density. We also find out the attributes of covariate shifts are ordered by corruption difficulty ranking (CDR) for the datasets. To the best of our knowledge, this is the first method to measure data corruption difficulty with generative models using Wasserstein Distance, Mutual Information and Minimal Description Length. In this paper, we pose interesting experimental results that the MNIST-C trained generative model is most deteriorated by fog, impulse noise and stripe corruption types. This can be interpreted that those attributes are challenging corruptions to the generative model in uncertainty and complexity. By training in-distribution data only, HYBOOD achieves out-of-distribution detection performance for distinguishable covariate and semantic shifts, and quantifying covariate shift ranking.
Catastrophic forgetting is when a neural network loses previously learnt information after learning a new task sequentially. Avoiding catastrophic forgetting could reduce the resources necessary to update neural networks. Recently, Kolmogorov–Arnold Networks (KAN) gained the community's attention as preliminary experiments suggest KAN avoid catastrophic forgetting. KAN replace neural network edges with learnable B-splines and sum incoming edges in nodes. Proponents of KAN argue they avoid forgetting, are more accurate, are interpretable, and use fewer parameters. Our work investigates the claims that KAN avoid catastrophic forgetting, finding that they fail to do so on more complex datasets containing features that overlap between tasks. We give a simple explanation as to why and how KAN catastrophically forget. Motivated by evidence suggesting KAN are superior for symbolic regression, we augment KAN in the same ways as multilayer perceptron (MLP) to perform continual learning tasks, making special accommodations to support KAN. Our experiments found that unmodified KAN often forget more than MLP, but KAN can be better than MLP when combined with continual learning strategies. We aim to highlight some of the current shortcomings and strengths associated with KAN for continual learning.
Balancing predictive power and interpretability has long been a challenging research area, particularly in powerful yet complex models like neural networks, where nonlinearity obstructs direct interpretation. This paper introduces a novel approach to constructing an explainable neural network that harmonizes predictiveness and explainability. Our model is designed as a linear combination of a sparse set of jointly learned features, each derived from a different trainable function applied to a single 1-dimensional input feature. Leveraging the ability to learn arbitrarily complex relationships, our neural network architecture enables automatic selection of a sparse set of important features, with the final prediction being a sum of rescaled versions of these features. We demonstrate the ability to select significant features while maintaining comparable predictive performance and direct interpretability through extensive experiments on synthetic and real-world datasets. We also provide theoretical analysis on the generalization bounds of our framework, which is favorably linear in the number of selected features and only logarithmic in the number of input features. We further lift any dependence of sample complexity on the number of parameters or the architectural details under very mild conditions. Our research paves the way for further research on sparse and explainable neural networks with guarantees.
With the rapid development of artificial intelligence (AI), especially in the medical field, the need for its explainability has grown. In medical image analysis, a high degree of transparency and model interpretability can help clinicians better understand and trust the decision-making process of AI models. In this study, we propose a Knowledge Distillation (KD) based approach that aims to enhance the transparency of the AI model in medical image analysis. The initial step is to use traditional CNN to obtain a teacher model and then use KD to simplify the CNN architecture, retain most of the features of the data set, and reduce the number of network layers. It also uses the feature map of the student model to perform hierarchical analysis to identify key features and decision-making processes. This leads to intuitive visual explanations. We selected three public medical data sets (brain tumor, eye disease, and Alzheimer's disease) to test our method. It shows that even when the number of layers is reduced, our model provides a remarkable result in the test set and reduces the time required for the interpretability analysis.
The applicability of drug molecules in various clinical scenarios is significantly influenced by a diverse range of molecular properties. By leveraging self-supervised conditions such as atom attributes and interatomic bonds, existing advanced molecular foundation models can generate expressive representations of these molecules. However, such models often overlook the fixed association patterns within molecules that influence physiological or chemical properties. In this paper, we introduce a novel association pattern-aware message passing method, which can serve as an effective yet general plug-and-play plugin, thereby enhancing the atom representations generated by molecular foundation models without requiring additional pretraining. Additionally, molecular property-specific pattern libraries are constructed to collect the generated interpretable common patterns that bind to these properties. Extensive experiments conducted on 11 benchmark molecular property prediction tasks across 8 advanced molecular foundation models demonstrate significant superiority of the proposed method, with performance improvements of up to approximately 20%. Furthermore, a property-specific pattern library is tailored for blood-brain barrier penetration, which has undergone corresponding mechanistic validation.
Extracting Interpretable Task-Specific Circuits from Large Language Models for Faster Inference
PDF ↗Large Language Models (LLMs) have shown impressive performance across a wide range of tasks. However, the size of LLMs is steadily increasing, hindering their application on computationally constrained environments. On the other hand, despite their general capabilities, there are many situations where only one specific task is performed, rendering all other capabilities unnecessary and wasteful. This leads us to the following question: Is it possible to extract the minimal subset from an LLM that is able to perform a specific task in a faster, standalone manner? Recent works on Mechanistic Interpretability (MI) have shown that specific tasks are performed by a localized subset of components, or circuit. However, current techniques used to identify the circuit cannot be used to extract it for its standalone usage. In this work, we propose a novel approach to automatically extract the subset of the LLM that properly performs a targeted task requiring no additional training and a small amount of data samples. We evaluate our approach on different tasks and show that the resulting models are (i) considerably smaller, reducing the number of parameters up to 82.77% and (ii) more interpretable, as they focus on the circuit that is used to carry out the specific task, and can therefore be understood using MI techniques.
Graph neural networks(GNNs) have been demonstrated to depend on whether the node effective information is sufficiently passing. Discrete curvature (Ricci curvature) is used to study graph connectivity and information propagation efficiency with a geometric perspective, and has been raised in recent years to explore the efficient message-passing structure of GNNs. However, most empirical studies are based on directly observed graph structures or heuristic topological assumptions, and lack in-depth exploration of underlying optimal information transport structures for downstream tasks. We suggest that graph curvature optimization is more in-depth and essential than directly rewiring or learning for graph structure with richer message-passing characterization and better information transport interpretability. From both graph geometry and information theory perspectives, we propose the novel Discrete Curvature Graph Information Bottleneck (CurvGIB) framework to optimize the information transport structure and learn better node representations simultaneously. CurvGIB advances the Variational Information Bottleneck (VIB) principle for Ricci curvature optimization to learn the optimal information transport pattern for specific downstream tasks. The learned Ricci curvature is used to refine the optimal transport structure of the graph, and the node representation is fully and efficiently learned. Moreover, for the computational complexity of Ricci curvature differentiation, we combine Ricci flow and VIB to deduce a curvature optimization approximation to form a tractable IB objective function. Extensive experiments on various datasets demonstrate the superior effectiveness and interpretability of CurvGIB.
Considering the difficulty of interpreting generative model output, there is significant current research focused on determining meaningful evaluation metrics. Several recent approaches utilize "precision" and "recall," borrowed from the classification domain, to individually quantify the output fidelity (realism) and output diversity (representation of the real data variation), respectively. With the increase in metric proposals, there is a need for a unifying perspective, allowing for easier comparison and clearer explanation of their benefits and drawbacks. To this end, we unify a class of kth-nearest neighbors (kNN)-based metrics under an information-theoretic lens using approaches from kNN density estimation. Additionally, we propose a tri-dimensional metric composed of Precision Cross-Entropy (PCE), Recall Cross-Entropy (RCE), and Recall Entropy (RE), which separately measure fidelity and two distinct aspects of diversity, inter- and intra-class. Our domain-agnostic metric, derived from the information-theoretic concepts of entropy and cross-entropy, can be dissected for both sample- and mode-level analysis. Our detailed experimental results demonstrate the sensitivity of our metric components to their respective qualities and reveal undesirable behaviors of other metrics.
Tree ensembles, including boosting methods, are highly effective and widely used for tabular data. However, large ensembles lack interpretability and require longer inference times. We introduce a method to prune a tree ensemble into a reduced version that is "functionally identical" to the original model. In other words, our method guarantees that the prediction function stays unchanged for any possible input. As a consequence, this pruning algorithm is lossless for any aggregated metric. We formalize the problem of functionally identical pruning on ensembles, introduce an exact optimization model, and provide a fast yet highly effective method to prune large ensembles. Our algorithm iteratively prunes considering a finite set of points, which is incrementally augmented using an adversarial model. In multiple computational experiments, we show that our approach provides a "free lunch", significantly reducing the ensemble size without altering the model's behavior. Thus, we can preserve state-of-the-art performance at a fraction of the original model's size.
Multi-view learning methods leverage multiple data sources to enhance perception by mining correlations across views, typically relying on predefined categories. However, deploying these models in real-world scenarios presents two primary openness challenges. 1) Lack of Interpretability: The integration mechanisms of multi-view data in existing black-box models remain poorly explained; 2) Insufficient Generalization: Most models are not adapted to multi-view scenarios involving unknown categories. To address these challenges, we propose OpenViewer, an openness-aware multi-view learning framework with theoretical support. This framework begins with a Pseudo-Unknown Sample Generation Mechanism to efficiently simulate open multi-view environments and previously adapt to potential unknown samples. Subsequently, we introduce an Expression-Enhanced Deep Unfolding Network to intuitively promote interpretability by systematically constructing functional prior-mapping modules and effectively providing a more transparent integration mechanism for multi-view data. Additionally, we establish a Perception-Augmented Open-Set Training Regime to significantly enhance generalization by precisely boosting confidences for known categories and carefully suppressing inappropriate confidences for unknown ones. Experimental results demonstrate that OpenViewer effectively addresses openness challenges while ensuring recognition performance for both known and unknown samples.
Sequential problems are ubiquitous in AI, such as in reinforcement learning or natural language processing. State-of-the-art deep sequential models, like transformers, excel in these settings but fail to guarantee the satisfaction of constraints necessary for trustworthy deployment. In contrast, neurosymbolic AI (NeSy) provides a sound formalism to enforce constraints in deep probabilistic models but scales exponentially on sequential problems. To overcome these limitations, we introduce relational neurosymbolic Markov models (NeSy-MMs), a new class of end-to-end differentiable sequential models that integrate and provably satisfy relational logical constraints. We propose a strategy for inference and learning that scales on sequential settings, and that combines approximate Bayesian inference, automated reasoning, and gradient estimation. Our experiments show that NeSy-MMs can solve problems beyond the current state-of-the-art in neurosymbolic AI and still provide strong guarantees with respect to desired properties. Moreover, we show that our models are more interpretable and that constraints can be adapted at test time to out-of-distribution scenarios.
In many real-world applications, data is inherently decentralized, necessitating data analysis methods that prioritize privacy while delivering interpretable results. Federated Non-Negative Matrix Factorization (FedNMF) meets this requirement by factorizing latent components from distributed data that cannot be freely shared among clients. A significant challenge in FedNMF arises when clients converge on different solutions due to prolonged independent optimization, leading to drift and incoherent models. While Federated Learning (FL) typically mitigates drift through frequent synchronizations and strong regularization, it often overlooks critical properties of Non-Negative Matrix Factorization, such as permutation invariance. As a result, solutions from FedNMF clients may be misidentified by FL drift as distinct, despite being equivalent. Using an alignment-aware drift, we create coherence through proximal optimization and barycenter aggregation for FedNMF. We analyze the computational complexity of our approach, provide efficient heuristics, and ensure the convergence of our algorithms. On a diverse set of real-world and synthetic datasets, we demonstrate the effectiveness of our methods.
Neural Reasoning Networks: Efficient Interpretable Neural Networks with Automatic Textual Explanations
PDF ↗Recent advances in machine learning have led to a surge in adoption of neural networks for various tasks, but lack of interpretability remains an issue for many others in which an understanding of the features influencing the prediction is necessary to ensure fairness, safety, and legal compliance. In this paper we consider one class of such tasks, tabular dataset classification, and propose a novel neuro-symbolic architecture, Neural Reasoning Networks (NRN), that is scalable and generates logically sound textual explanations for its predictions. NRNs are connected layers of logical neurons that implement a form of real valued logic. A training algorithm (R-NRN) learns the weights of the network as usual using gradient descent optimization with backprop, but also learns the network structure itself using a bandit-based optimization. Both are implemented in an extension to PyTorch that takes full advantage of GPU scaling and batched training. Evaluation on a diverse set of 22 open-source datasets for tabular classification demonstrates performance (measured by ROC AUC) which improves over Multilayer Perceptron (MLP) and is statistically similar to other state-of-the-art approaches such as Random Forest, XGBoost and Gradient Boosted Trees, while offering 43% faster training and a more than 2 orders of magnitude reduction in the number of parameters required, on average. Furthermore, R-NRN explanations are shorter than the compared approaches while producing more accurate feature importance scores.
Counterfactual explanations are one of the prominent eXplainable Artificial Intelligence (XAI) techniques, and suggest changes to input data that could alter predictions, leading to more favourable outcomes. Existing counterfactual methods do not readily apply to temporal domains, such as that of process mining, where data take the form of traces of activities that must obey to temporal background knowledge expressing which dynamics are possible and which not. Specifically, counterfactuals generated off-the-shelf may violate the background knowledge, leading to inconsistent explanations. This work tackles this challenge by introducing a novel approach for generating temporally constrained counterfactuals, guaranteed to comply by design with background knowledge expressed in Linear Temporal Logic on process traces (LTLp). We do so by infusing automata-theoretic techniques for LTLp inside a genetic algorithm for counterfactual generation. The empirical evaluation shows that the generated counterfactuals are temporally meaningful and more interpretable for applications involving temporal dependencies.