Generating high-fidelity synthetic tabular data remains a critical challenge for enhancing data availability in privacy-sensitive and lowresource domains.Recent approaches leverage LLMs by representing table rows as sequences, yet suffer from two fundamental limitations:(1) they model feature dependencies densely, introducing spurious correlations; and (2) they assume static relationships between features, ignoring how these dependencies vary with feature values.To overcome these limitations, we introduce SAGE (Sparse Adaptive Guidance), a novel LLM-based generation framework that enforces sparse and dynamic dependency guidance.SAGE discretizes features into value-aware pseudo-features and constructs a mutual information-based sparse dependency graph.This graph adaptively guides generation through explicit context selection or implicit logit correction, enabling LLMs to focus on truly relevant information during synthesis.Our extensive experiments across six datasets and multiple tasks reveal that SAGE not only improves data fidelity and downstream utility, boosting F1 scores by 10% compared to previous LLM-based methods, but also reduces policy violations by one point.These results highlight the importance of adaptive structure in tabular data generation and provide new insights into context-sensitive control of LLMs. 1 * Equal contribution. 1 Our code is publicly available at https://github.com/ ShuoYangtum/SAGE.

Large language models often display heterogeneous moral preferences across settings.We study inference-time steering toward a desired ethical framework while preserving general competence.We present Convergent-Divergent Routing, which traces and edits minimal branch points inside transformer blocks where ethicalframework-related pathways first converge and then diverge.Gating non-target branches at these loci blocks the downstream propagation while leaving upstream computations intact.We find that this intervention alone increases targeted ethical-framework reasoning.To achieve fine-grained control, we adapt Common Spatial Patterns to the residual stream and extract, for each branch-point layer, a pair of directions that discriminate between utilitarian and deontological frameworks.We then introduce Dual Logit Calibration, a closed-form, minimum-ℓ 2 -norm update that moves the residual within this two-dimensional subspace so the resulting directional projections align with userspecified preference weights.Experiments on real-life moral dilemmas show that our method reliably achieves preference calibration and largely preserves general capabilities, outperforming recent baselines while providing an interpretable mechanism.

Despite the widespread use of LLMs due to their superior performance in various tasks, their high computational costs often lead potential users to opt for the pretraining-finetuning pipeline. However, biases prevalent in manually constructed datasets can introduce spurious correlations between tokens and labels, creating so-called shortcuts and hindering the generalizability of fine-tuned models. Existing debiasing methods often rely on prior knowledge of specific dataset biases, which is challenging to acquire a priori. We propose RAZOR (Rewriting And Zero-bias Optimization Refinement), a novel, unsupervised, and data-focused debiasing approach based on text rewriting for shortcut mitigation. RAZOR leverages LLMs to iteratively rewrite potentially biased text segments by replacing them with heuristically selected alternatives in a shortcut space defined by token statistics and positional information. This process aims to align surface-level text features more closely with diverse label distributions, thereby promoting the learning of genuine linguistic patterns. Compared with unsupervised SoTA models, RAZOR improves by 3.5% on the FEVER and 6.5% on MNLI and SNLI datasets according to the F1 score. Additionally, RAZOR effectively mitigates specific known biases, reducing bias-related terms by x2 without requiring prior bias information, a result that is on par with SoTA models that leverage prior information. Our work prioritizes data manipulation over architectural modifications, emphasizing the pivotal role of data quality in enhancing model performance and fairness. This research contributes to developing more robust evaluation benchmarks for debiasing methods by incorporating metrics for bias reduction and overall model efficacy.

Tabular data is critical across diverse domains, yet high-quality datasets remain scarce due to privacy concerns and the cost of collection.Contemporary approaches adopt large language models (LLMs) for tabular augmentation, but exhibit two major limitations: (1) dense dependency modeling among tabular features that can introduce bias, and (2) high computational overhead in sampling.To address these issues, we propose SPADA (for SPArse Dependencydriven Augmentation), a lightweight generative framework that explicitly captures sparse dependencies via an LLM-induced graph.We treat each feature as a node and synthesize values by traversing the graph, conditioning each feature solely on its parent nodes.We explore two synthesis strategies: a non-parametric method using Gaussian kernel density estimation, and a conditional normalizing flow model that learns invertible mappings for conditional density estimation.Experiments on four datasets show that SPADA reduces constraint violations by 4% compared to diffusion-based methods and accelerates generation by nearly 9,500× over LLM-based baselines.1

Transformers have achieved great success in numerous NLP tasks but continue to exhibit notable gaps in multi-step factual reasoning, especially when real-world knowledge is sparse. Recent advances in grokking have demonstrated that neural networks can transition from memorizing to perfectly generalizing once they detect underlying logical patterns – yet these studies have primarily used small, synthetic tasks. In this paper, for the first time, we extend grokking to real-world factual data and address the challenge of dataset sparsity by augmenting existing knowledge graphs with carefully designed synthetic data to raise the ratio $\phi_r$ of inferred facts to atomic facts above the threshold required for grokking. Surprisingly, we find that even factually incorrect synthetic data can strengthen emergent reasoning circuits rather than degrade accuracy, as it forces the model to rely on relational structure rather than memorization. When evaluated on multi-hop reasoning benchmarks, our approach achieves up to 95–100% accuracy on 2WikiMultiHopQA – substantially improving over strong baselines and matching or exceeding current state-of-the-art results. We further provide an in-depth analysis of how increasing $\phi_r$ drives the formation of generalizing circuits inside Transformers. Our findings suggest that grokking-based data augmentation can unlock implicit multi-hop reasoning capabilities, opening the door to more robust and interpretable factual reasoning in large-scale language models.

Counterfactual explainability seeks to uncover model decisions by identifying minimal changes to the input that alter the predicted outcome. This task becomes particularly challenging for graph data due to preserving structural integrity and semantic meaning. Unlike prior approaches that rely on forward perturbation mechanisms, we introduce Graph Inverse Style Transfer (GIST), the first framework to re-imagine graph counterfactual generation as a backtracking process, leveraging spectral style transfer. By aligning the global structure with the original input spectrum and preserving local content faithfulness, GIST produces valid counterfactuals as interpolations between the input style and counterfactual content. Tested on 8 binary and multi-class graph classification benchmarks, GIST achieves a remarkable +7.6% improvement in the validity of produced counterfactuals and significant gains (+45.5%) in faithfully explaining the true class distribution. Additionally, GIST’s backtracking mechanism effectively mitigates overshooting the underlying predictor’s decision boundary, minimizing the spectral differences between the input and the counterfactuals. These results challenge traditional forward perturbation methods, offering a novel perspective that advances graph explainability.

Shortcut learning undermines model generalization to out-of-distribution data. While the literature attributes shortcuts to biases in superficial features, we show that imbalances in the semantic distribution of sample embeddings induce spurious semantic correlations, compromising model robustness. To address this issue, we propose SCISSOR (Semantic Cluster Intervention for Suppressing ShORtcut), a Siamese network-based debiasing approach that remaps the semantic space by discouraging latent clusters exploited as shortcuts. Unlike prior data-debiasing approaches, SCISSOR eliminates the need for data augmentation and rewriting. We evaluate SCISSOR on 6 models across 4 benchmarks: Chest-XRay and Not-MNIST in computer vision, and GYAFC and Yelp in NLP tasks. Compared to several baselines, SCISSOR reports +5.3 absolute points in F1 score on GYAFC, +7.3 on Yelp, +7.7 on Chest-XRay, and +1 on Not-MNIST. SCISSOR is also highly advantageous for lightweight models with $\tilde$9.5% improvement on F1 for ViT on computer vision datasets and $\tilde$11.9% for BERT on NLP. Our study redefines the landscape of model generalization by addressing overlooked semantic biases, establishing SCISSOR as a foundational framework for mitigating shortcut learning and fostering more robust, bias-resistant AI systems.

It is commonly assumed that digital learning environments such as intelligent tutoring systems facilitate learning and positively impact achievement. This study explores how different groups of students exhibit distinct relationships between learning behaviors and academic achievement in an intelligent tutoring system for English as a foreign language. We examined whether these differences are linked to students’ prior knowledge, personality traits, and motivation. We collected behavioral trace data from 507 German seventh-grade students during the 2021/22 school year and applied machine learning models to predict English performance based on learning behaviors (best-performing model’s 2 = .41). To understand the impact of specific behaviors, we applied the explainable AI method SHAP and identified three student clusters with distinct learning behavior patterns. Subsequent analyses revealed that these clusters also varied in prior knowledge and motivation: one with high prior knowledge and average motivation, another with low prior knowledge and average motivation, and a third with both low prior knowledge and low motivation. Our findings suggest that learning behaviors are linked differently to academic success across students and are closely tied to their prior knowledge and motivation. This hints towards the importance of personalizing learning systems to support individual learning needs better.

With recent advances in machine learning and big data, it is now possible to create synthetic images that look real. Face generation is often of particular interest, as faces can be used for various purposes. However, improper use of such content can lead to the dissemination of false information, such as fake news, and thus pose a threat to society. This work studies whether people believe the truthfulness of faces using eye tracking and self-reports, including free-form textual explanations when participants encounter real and computer-generated faces. We used three different datasets for our evaluations, and our experimental results show that while people are relatively better at identifying the truthfulness of real faces and faces generated by earlier machine learning algorithms with different gazing behaviors in viewing and rating phases, they perform less accurately when deciding the truthfulness of synthetic face images that are generated by newer algorithms. Our findings provide important insights for society and policymakers.

We examine machine learning models in a setup where individuals have the choice to share optional personal information with a decision-making system, as seen in modern insurance pricing models. Some users consent to their data being used whereas others object and keep their data undisclosed. In this work, we show that the decision not to share data can be considered as information in itself that should be protected to respect users' privacy. This observation raises the overlooked problem of how to ensure that users who protect their personal data do not suffer any disadvantages as a result. To address this problem, we formalize protection requirements for models which only use the information for which active user consent was obtained. This excludes implicit information contained in the decision to share data or not. We offer the first solution to this problem by proposing the notion of Protected User Consent (PUC), which we prove to be loss-optimal under our protection requirement. We observe that privacy and performance are not fundamentally at odds with each other and that it is possible for a decision maker to benefit from additional data while respecting users' consent. To learn PUC-compliant models, we devise a model-agnostic data augmentation strategy with finite sample convergence guarantees. Finally, we analyze the implications of PUC on challenging real datasets, tasks, and models.

To address bias issues, fair machine learning usually jointly optimizes two (or more) metrics aiming at predictive utility and fairness. However, the inherent under-representation of minorities in the data often makes the disparate impact of subpopulations less noticeable and difficult to deal with during learning. In this paper, we propose a novel adversarial reweighting method to address such disparate impact. To balance the data distribution between the majority and the minority groups, our approach prefers samples from the majority group that are closer to the minority group as evaluated by the Wasserstein distance. Theoretical analysis shows the effectiveness of our adversarial reweighting approach. Experiments demonstrate that our approach mitigates disparate impact without sacrificing classification accuracy, outperforming related state-of-the-art methods on image and tabular benchmark datasets. Code is available at https://github.com/zhaoxuan00707/wasserstein_reweight.

Wide usage of ChatGPT has highlighted the potential of reinforcement learning from human feedback. However, its training pipeline relies on manual ranking, a resource-intensive process. To reduce labor costs, we propose a self-supervised text ranking approach for applying Proximal-Policy-Optimization to fine-tune language models while eliminating the need for human annotators. Our method begins with probabilistic sampling to encourage a language model to generate diverse responses for each input. We then employ TextRank and ISODATA algorithms to rank and cluster these responses based on their semantics. Subsequently, we construct a reward model to learn the rank and optimize our generative policy. Our experimental results, conducted using two language models on three tasks, demonstrate that the models trained by our method considerably outperform baselines regarding BLEU, GLEU, and METEOR scores. Furthermore, our manual evaluation shows that our ranking results exhibit a remarkably high consistency with that of humans. This research significantly reduces training costs of proximal policy-guided models and demonstrates the potential for self-correction of language models.

We introduce an innovative approach to enhancing the empirical risk minimization (ERM) process in model training through a refined reweighting scheme of the training data to enhance fairness. This scheme aims to uphold the sufficiency rule in fairness by ensuring that optimal predictors maintain consistency across diverse sub-groups. We employ a bilevel formulation to address this challenge, wherein we explore sample reweighting strategies. Unlike conventional methods that hinge on model size, our formulation bases generalization complexity on the space of sample weights. We discretize the weights to improve training speed. Empirical validation of our method showcases its effectiveness and robustness, revealing a consistent improvement in the balance between prediction performance and fairness metrics across various experiments. Code is available at https://github.com/zhaoxuan00707/Reweighting_for_sufficiency.

Summarizing long pieces of text is a principal task in natural language processing with Machine Learning-based text generation models such as Large Language Models (LLM) being particularly suited to it.Yet these models are often used as black-boxes, making them hard to interpret and debug.This has led to calls by practitioners and regulatory bodies to improve the explainability of such models as they find ever more practical use.In this survey, we present a dual-perspective review of the intersection between explainability and summarization by reviewing the current state of explainable text summarization and also highlighting how summarization techniques are effectively employed to improve explanations.

We present GRACIE (Graph Recalibration and Adaptive Counterfactual Inspection and Explanation), a novel approach for generative classification and counterfactual explanations of dynamically changing graph data. We study graph classification problems through the lens of generative classifiers. We propose a dynamic, self-supervised latent variable model that updates by identifying plausible counterfactuals for input graphs and recalibrating decision boundaries through contrastive optimization. Unlike prior work, we do not rely on linear separability between the learned graph representations to find plausible counterfactuals. Moreover, GRACIE eliminates the need for stochastic sampling in latent spaces and graph-matching heuristics. Our work distills the implicit link between generative classification and loss functions in the latent space, a key insight to understanding recent successes with this architecture. We further observe the inherent trade-off between validity and pulling explainee instances towards the central region of the latent space, empirically demonstrating our theoretical findings. In extensive experiments on synthetic and real-world graph data, we attain considerable improvements, reaching ~99% validity when sampling sets of counterfactuals even in the challenging setting of dynamic data landscapes.

Explainable AI (XAI) is widely viewed as a sine qua non for ever-expanding AI research. A better understanding of the needs of XAI users, as well as human-centered evaluations of explainable models are both a necessity and a challenge. In this paper, we explore how human-computer interaction (HCI) and AI researchers conduct user studies in XAI applications based on a systematic literature review. After identifying and thoroughly analyzing 97 core papers with human-based XAI evaluations over the past five years, we categorize them along the measured characteristics of explanatory methods, namely trust, understanding, usability, and human-AI collaboration performance. Our research shows that XAI is spreading more rapidly in certain application domains, such as recommender systems than in others, but that user evaluations are still rather sparse and incorporate hardly any insights from cognitive or social sciences. Based on a comprehensive discussion of best practices, i.e., common models, design choices, and measures in user studies, we propose practical guidelines on designing and conducting user studies for XAI researchers and practitioners. Lastly, this survey also highlights several open research directions, particularly linking psychological science and human-centered XAI.

Heterogeneous tabular data are the most commonly used form of data and are essential for numerous critical and computationally demanding applications. On homogeneous datasets, deep neural networks have repeatedly shown excellent performance and have therefore been widely adopted. However, their adaptation to tabular data for inference or data generation tasks remains highly challenging. To facilitate further progress in the field, this work provides an overview of state-of-the-art deep learning methods for tabular data. We categorize these methods into three groups: data transformations, specialized architectures, and regularization models. For each of these groups, our work offers a comprehensive overview of the main approaches. Moreover, we discuss deep learning approaches for generating tabular data and also provide an overview over strategies for explaining deep models on tabular data. Thus, our first contribution is to address the main research streams and existing methodologies in the mentioned areas while highlighting relevant challenges and open research questions. Our second contribution is to provide an empirical comparison of traditional machine learning methods with 11 deep learning approaches across five popular real-world tabular datasets of different sizes and with different learning objectives. Our results, which we have made publicly available as competitive benchmarks, indicate that algorithms based on gradient-boosted tree ensembles still mostly outperform deep learning models on supervised learning tasks, suggesting that the research progress on competitive deep learning models for tabular data is stagnating. To the best of our knowledge, this is the first in-depth overview of deep learning approaches for tabular data; as such, this work can serve as a valuable starting point to guide researchers and practitioners interested in deep learning with tabular data.

In recent years, game-theoretic Shapley values have gained increasing attention with respect to local model explanation by feature attributions. While the approach using Shapley values is model-independent, their (exact) computation is usually intractable, so efficient model-specific algorithms have been devised including approaches for decision trees or their ensembles in general. Our work goes further in this direction by extending the interventional TreeSHAP algorithm to piecewise linear regression trees, which gained more attention in the past few years. To this end, we introduce a decomposition of the contribution function based on decision paths, which allows a more comprehensible formulation of SHAP algorithms for tree-based models. Our algorithm can also be readily applied to computing SHAP interaction values of these models. In particular, as the main contribution of this paper, we provide a more efficient approach of interventional SHAP for tree-based models by precomputing statistics of the background data based on the tree structure.

The importance of achieving fairness in machine learning models cannot be overstated. Recent research has pointed out that fairness should be examined from a causal perspective, and several fairness notions based on the on Pearl’s causal framework have been proposed. In this paper, we construct a reweighting scheme of datasets to address causal fairness. Our approach aims at mitigating bias by considering the causal relationships among variables and incorporating them into the reweighting process. The proposed method adopts two neural networks, whose structures are intentionally used to reflect the structures of a causal graph and of an interventional graph. The two neural networks can approximate the causal model of the data, and the causal model of interventions. Furthermore, reweighting guided by a discriminator is applied to achieve various fairness notions. Experiments on real-world datasets show that our method can achieve causal fairness on the data while remaining close to the original data for downstream tasks.

We propose a novel and practical privacy notion called $f$-Membership Inference Privacy ($f$-MIP), which explicitly considers the capabilities of realistic adversaries under the membership inference attack threat model. Consequently, $f$-MIP offers interpretable privacy guarantees and improved utility (e.g., better classification accuracy). In particular, we derive a parametric family of $f$-MIP guarantees that we refer to as $\mu$-Gaussian Membership Inference Privacy ($\mu$-GMIP) by theoretically analyzing likelihood ratio-based membership inference attacks on stochastic gradient descent (SGD). Our analysis highlights that models trained with standard SGD already offer an elementary level of MIP. Additionally, we show how $f$-MIP can be amplified by adding noise to gradient updates. Our analysis further yields an analytical membership inference attack that offers two distinct advantages over previous approaches. First, unlike existing state-of-the-art attacks that require training hundreds of shadow models, our attack does not require any shadow model. Second, our analytical attack enables straightforward auditing of our privacy notion $f$-MIP. Finally, we quantify how various hyperparameters (e.g., batch size, number of model parameters) and specific data characteristics determine an attacker's ability to accurately infer a point's membership in the training set. We demonstrate the effectiveness of our method on models trained on vision and tabular datasets.

Interest in understanding and factorizing learned embedding spaces through conceptual explanations is steadily growing. When no human concept labels are available, concept discovery methods search trained embedding spaces for interpretable concepts like object shape or color that can provide post-hoc explanations for decisions. Unlike previous work, we argue that concept discovery should be identifiable, meaning that a number of known concepts can be provably recovered to guarantee reliability of the explanations. As a starting point, we explicitly make the connection between concept discovery and classical methods like Principal Component Analysis and Independent Component Analysis by showing that they can recover independent concepts under non-Gaussian distributions. For dependent concepts, we propose two novel approaches that exploit functional compositionality properties of image-generating processes. Our provably identifiable concept discovery methods substantially outperform competitors on a battery of experiments including hundreds of trained models and dependent concepts, where they exhibit up to 29 % better alignment with the ground truth. Our results highlight the strict conditions under which reliable concept discovery without human labels can be guaranteed and provide a formal foundation for the domain. Our code is available online.

With increasing digitalization, Artificial Intelligence (AI) is becoming ubiquitous. AI-based systems to identify, optimize, automate, and scale solutions to complex economic and societal problems are being proposed and implemented. This has motivated regulation efforts, including the Proposal of an EU AI Act. This interdisciplinary position paper considers various concerns surrounding fairness and discrimination in AI, and discusses how AI regulations address them, focusing on (but not limited to) the Proposal. We first look at AI and fairness through the lenses of law, (AI) industry, sociotechnology, and (moral) philosophy, and present various perspectives. Then, we map these perspectives along three axes of interests: (i) Standardization vs. Localization, (ii) Utilitarianism vs. Egalitarianism, and (iii) Consequential vs. Deontological ethics which leads us to identify a pattern of common arguments and tensions between these axes. Positioning the discussion within the axes of interest and with a focus on reconciling the key tensions, we identify and propose the roles AI Regulation should take to make the endeavor of the AI Act a success in terms of AI fairness concerns.

As complex machine learning models are increasingly used in sensitive applications like banking, trading or credit scoring, there is a growing demand for reliable explanation mechanisms. Local feature attribution methods have become a popular technique for post-hoc and model-agnostic explanations. However, attribution methods typically assume a stationary environment in which the predictive model has been trained and remains stable. As a result, it is often unclear how local attributions behave in realistic, constantly evolving settings such as streaming and online applications. In this paper, we discuss the impact of temporal change on local feature attributions. In particular, we show that local attributions can become obsolete each time the predictive model is updated or concept drift alters the data generating distribution. Consequently, local feature attributions in data streams provide high explanatory power only when combined with a mechanism that allows us to detect and respond to local changes over time. To this end, we present CDLEEDS, a flexible and model-agnostic framework for detecting local change and concept drift. CDLEEDS serves as an intuitive extension of attribution-based explanation techniques to identify outdated local attributions and enable more targeted recalculations. In experiments, we also show that the proposed framework can reliably detect both local and global concept drift. Accordingly, our work contributes to a more meaningful and robust explainability in online machine learning.

With the increasing use of the Internet, people encounter a variety of news in online media and social media every day. For digital content without fact-checking mechanisms, it is likely that people perceive fake news as real when they do not have extensive knowledge about the news topic. In this paper, we study human eye movements when reading fake news and real news. Our results suggest that people regress more with their eyes when reading fake news, while the time until the first fixation in the text area of interest is not a distinguishing factor between real and fake content. Our results show that although the truthfulness of the content is not known to people in advance, their visual behavior differs when reading such content, indicating a higher level of confusion when reading fake content.

Data streams are ubiquitous in modern business and society. In practice, data streams may evolve over time and cannot be stored indefinitely. Effective and transparent machine learning on data streams is thus often challenging. Hoeffding Trees have emerged as a state-of-the art for online predictive modelling. They are easy to train and provide meaningful convergence guarantees under a stationary process. Yet, at the same time, Hoeffding Trees often require heuristic and costly extensions to adjust to distributional change, which may considerably impair their interpretability. In this work, we revisit Model Trees for machine learning in evolving data streams. Model Trees are able to maintain more flexible and locally robust representations of the active data concept, making them a natural fit for data stream applications. Our novel framework, called Dynamic Model Tree, satisfies desirable consistency and minimality properties. In experiments with synthetic and real-world tabular streaming data sets, we show that the proposed framework can drastically reduce the number of splits required by existing incremental decision trees. At the same time, our framework often outperforms state-of-the-art models in terms of predictive quality - especially when concept drift is involved. Dynamic Model Trees are thus a powerful online learning framework that contributes to more lightweight and interpretable machine learning in data streams.

With a variety of local feature attribution methods being proposed in recent years, follow-up work suggested several evaluation strategies. To assess the attribution quality across different attribution techniques, the most popular among these evaluation strategies in the image domain use pixel perturbations. However, recent advances discovered that different evaluation strategies produce conflicting rankings of attribution methods and can be prohibitively expensive to compute. In this work, we present an information-theoretic analysis of evaluation strategies based on pixel perturbations. Our findings reveal that the results are strongly affected by information leakage through the shape of the removed pixels as opposed to their actual values. Using our theoretical insights, we propose a novel evaluation framework termed Remove and Debias (ROAD) which offers two contributions: First, it mitigates the impact of the confounders, which entails higher consistency among evaluation strategies. Second, ROAD does not require the computationally expensive retraining step and saves up to 99% in computational costs compared to the state-of-the-art. We release our source code at https://github.com/tleemann/road_evaluation.

Local approximations are popular methods to scale Gaussian processes (GPs) to big data. Local approximations reduce time complexity by dividing the original dataset into subsets and training a local expert on each subset. Aggregating the experts' prediction is done assuming either conditional dependence or independence between the experts. Imposing the conditional independence assumption (CI) between the experts renders the aggregation of different expert predictions time efficient at the cost of poor uncertainty quantification. On the other hand, modeling dependent experts can provide precise predictions and uncertainty quantification at the expense of impractically high computational costs. By eliminating weak experts via a theory-guided expert selection step, we substantially reduce the computational cost of aggregating dependent experts while ensuring calibrated uncertainty quantification. We leverage techniques from the literature on undirected graphical models, using sparse precision matrices that encode conditional dependencies between experts to select the most important experts. Moreover, our approach also provides a solution to the poor uncertainty quantification in CI-based models.

We present TEyeD, the world’s largest unified public data set of eye images taken with head-mounted devices. TEyeD was acquired with seven different head-mounted eye trackers. Among them, two eye trackers were integrated into virtual reality (VR) or augmented reality (AR) devices. The images in TEyeD were obtained from various tasks, including car rides, simulator rides, outdoor sports activities, and daily indoor activities. The data set includes 2D&3D landmarks, semantic segmentation, 3D eyeball annotation and the gaze vector and eye movement types for all images. Landmarks and semantic segmentation are provided for the pupil, iris and eyelids. Video lengths vary from a few minutes to several hours. With more than 20 million carefully annotated images, TEyeD provides a unique, coherent resource and a valuable foundation for advancing research in the field of computer vision, eye tracking and gaze estimation in modern VR and AR applications. Data and code at DOWNLOAD LINK.

We present an alternative layer to convolution layers in convolutional neural networks (CNNs). Our approach reduces the complexity of convolutions by replacing it with binary decisions. Those binary decisions are used as indexes to conditional distributions where each weight represents a leaf in a decision tree. This means that only the indices to the weights need to be determined once, thus reducing the complexity of convolutions by the depth of the output tensor. Index computation is performed by simple binary decisions that require fewer cycles compared to conventionally used multiplications. In addition, we show how convolutions can be replaced by binary decisions. These binary decisions form indices in the conditional distributions and we show how they are used to replace 2D weight matrices as well as 3D weight tensors. These new layers can be trained like convolution layers in CNNs based on the backpropagation algorithm, for which we provide a formalization. Our results on multiple publicly available data sets show that our approach performs similar to conventional neuronal networks. Beyond the formalized reduction of complexity and the improved qualitative performance, we show the runtime improvement empirically compared to convolution layers.

The visual scanpath describes the shift of visual attention over time. Characteristic patterns in the attention shifts allow inferences about cognitive processes, performed tasks, intention, or expertise. To analyse such patterns, the scanpath is often represented as a sequence of symbols that can be used to calculate a similarity score to other scanpaths. However, as the length of the scanpath or the number of possible symbols increases, established methods for scanpath similarity become inefficient, both in terms of runtime and memory consumption. We present a MinHash approach for efficient scanpath similarity calculation. Our approach shows competitive results in clustering and classification of scanpaths compared to established methods such as Needleman-Wunsch, but at a fraction of the required runtime. Furthermore, with time complexity of and constant memory consumption, our approach is ideally suited for real-time operation or analyzing large amounts of data.

Data distributions in streaming environments are usually not stationary. In order to maintain a high predictive quality at all times, online learning models need to adapt to distributional changes, which are known as concept drift. The timely and robust identification of concept drift can be difficult, as we never have access to the true distribution of streaming data. In this work, we propose a novel framework for the detection of real concept drift, called ERICS. By treating the parameters of a predictive model as random variables, we show that concept drift corresponds to a change in the distribution of optimal parameters. To this end, we adopt common measures from information theory. The proposed framework is completely model-agnostic. By choosing an appropriate base model, ERICS is also capable to detect concept drift at the input level, which is a significant advantage over existing approaches. An evaluation on several synthetic and real-world data sets suggests that the proposed framework identifies concept drift more effectively and precisely than various existing works.

Distributed Gaussian processes (DGPs) are prominent local approximation methods to scale Gaussian processes (GPs) to large datasets. Instead of a global estimation, they train local experts by dividing the training set into subsets, thus reducing the time complexity. This strategy is based on the conditional independence assumption, which basically means that there is a perfect diversity between the local experts. In practice, however, this assumption is often violated, and the aggregation of experts leads to sub-optimal and inconsistent solutions. In this paper, we propose a novel approach for aggregating the Gaussian experts by detecting strong violations of conditional independence. The dependency between experts is determined by using a Gaussian graphical model, which yields the precision matrix. The precision matrix encodes conditional dependencies between experts and is used to detect strongly dependent experts and construct an improved aggregation. Using both synthetic and real datasets, our experimental evaluations illustrate that our new method outperforms other state-of-the-art (SOTA) DGP approaches while being substantially more time-efficient than SOTA approaches, which build on independent experts.

Feature selection can be a crucial factor in obtaining robust and accurate predictions. Online feature selection models, however, operate under considerable restrictions; they need to efficiently extract salient input features based on a bounded set of observations, while enabling robust and accurate predictions. In this work, we introduce FIRES, a novel framework for online feature selection. The proposed feature weighting mechanism leverages the importance information inherent in the parameters of a predictive model. By treating model parameters as random variables, we can penalize features with high uncertainty and thus generate more stable feature sets. Our framework is generic in that it leaves the choice of the underlying model to the user. Strikingly, experiments suggest that the model complexity has only a minor effect on the discriminative power and stability of the selected feature sets. In fact, using a simple linear model, FIRES obtains feature sets that compete with state-of-the-art methods, while dramatically reducing computation time. In addition, experiments show that the proposed framework is clearly superior in terms of feature selection stability.

Counterfactual explanations are usually obtainedby identifying the smallest change made to an input to change a prediction made by a fixed model (hereafter called sparse methods). Recent work, however, has revitalized an old insight: there often does not exist one superior solution to a prediction problem with respect to commonly used measures of interest (e.g. error rate). In fact, often multiple different classifiers give almost equal solutions. This phenomenon is known as predictive multiplicity (Breiman, 2001; Marx et al., 2019). In this work, we derive a general upper bound for the costs of counterfactual explanations under predictive multiplicity. Most notably, it depends on a discrepancy notion between two classifiers, which describes how differently they treat negatively predicted individuals. We then compare sparse and data support approaches empirically on real-world data. The results show that data support methods are more robust to multiplicity of different models. At the same time, we show that those methods have provably higher cost of generating counterfactual explanations under one fixed model. In summary, our theoretical and empirical results challenge the commonly held view that counterfactual recommendations should be sparse in general.

Counterfactual explanations can be obtained by identifying the smallest change made to an input vector to influence a prediction in a positive way from a user’s viewpoint; for example, from ’loan rejected’ to ’awarded’ or from ’high risk of cardiovascular disease’ to ’low risk’. Previous approaches would not ensure that the produced counterfactuals be proximate (i.e., not local outliers) and connected to regions with substantial data density (i.e., close to correctly classified observations), two requirements known as counterfactual faithfulness. Our contribution is twofold. First, drawing ideas from the manifold learning literature, we develop a framework, called C-CHVAE, that generates faithful counterfactuals. Second, we suggest to complement the catalog of counterfactual quality measures using a criterion to quantify the degree of difficulty for a certain counterfactual suggestion. Our real world experiments suggest that faithful counterfactuals come at the cost of higher degrees of difficulty.

Machine learning algorithms aim at minimizing the number of false decisions and increasing the accuracy of predictions. However, the high predictive power of advanced algorithms comes at the costs of transparency. State-of-the-art methods, such as neural networks and ensemble methods, result in highly complex models with little transparency. We propose shallow model trees as a way to combine simple and highly transparent predictive models for higher predictive power without losing the transparency of the original models. We present a novel split criterion for model trees that allows for significantly higher predictive power than state-of-the-art model trees while maintaining the same level of simplicity. This novel approach finds split points which allow the underlying simple models to make better predictions on the corresponding data. In addition, we introduce multiple mechanisms to increase the transparency of the resulting trees.

We address the problem of latent truth discovery, LTD for short, where the goal is to discover the underlying true values of entity attributes in the presence of noisy, conflicting or incomplete information. Despite a multitude of algorithms addressing the LTD problem, only little is known about their overall performance with respect to effectiveness, efficiency and robustness. The LTD model proposed in this paper is based on Restricted Boltzmann Machines, thus coined LTD-RBM. In extensive experiments on various heterogeneous and publicly available datasets, LTD-RBM is superior to state-of-the-art LTD techniques in terms of an overall consideration of effectiveness, efficiency and robustness.

In recent years artificial neural networks have become the method of choice for many pattern recognition tasks. Despite their overwhelming success, a rigorous and easy to interpret mathematical explanation of the influence of input variables on a output produced by a neural network is still missing.

Duplicates in a dataset are multiple representations of the same real-world entity and constitute a major data quality problem. This paper investigates the problem of estimating the number and sizes of duplicate record clusters in advance and describes a sampling-based method for solving this problem. In extensive experiments, on multiple datasets, we show that the proposed method reliably estimates the number of duplicate clusters, while being highly efficient.

In many applications involving scanpath analysis, especially when dynamic scenes are viewed, consecutive fixations and saccades, have to be identified and extracted from raw eye-tracking data in an online fashion. Since probabilistic methods can adapt not only to the individual viewing behavior, but also to changes in the scene, they are best suited for such tasks.

The Internet has enabled the creation of a growing number of large-scale knowledge bases in a variety of domains containing complementary information. Tools for automatically aligning these knowledge bases would make it possible to unify many sources of structured knowledge and answer complex queries. However, the efficient alignment of large-scale knowledge bases still poses a considerable challenge. Here, we present Simple Greedy Matching (SiGMa), a simple algorithm for aligning knowledge bases with millions of entities and facts. SiGMa is an iterative propagation algorithm that leverages both the structural information from the relationship graph and flexible similarity measures between entity properties in a greedy local search, which makes it scalable. Despite its greedy nature, our experiments indicate that SiGMa can efficiently match some of the world's largest knowledge bases with high accuracy. We provide additional experiments on benchmark datasets which demonstrate that SiGMa can outperform state-of-the-art approaches both in accuracy and efficiency.

Large amounts of graph-structured data are emerging from various avenues, ranging from natural and life sciences to social and semantic web communities. We address the problem of discovering subgraphs of entities that reflect latent topics in graph-structured data. These topics are structured meta-information providing further insights into the data. The presented approach effectively detects such topics by exploiting only the structure of the underlying graph, thus avoiding the dependency on textual labels, which are a scarce asset in prevalent graph datasets. The viability of our approach is demonstrated in experiments on real-world datasets.

The task of automatically tracking the visual attention in dynamic visual scenes is highly challenging. To approach it, we propose a Bayesian online learning algorithm. As the visual scene changes and new objects appear, based on a mixture model, the algorithm can identify and tell visual saccades (transitions) from visual fixation clusters (regions of interest). The approach is evaluated on real-world data, collected from eye-tracking experiments in driving sessions.

The prediction accuracy of any learning algorithm highly depends on the quality of the selected features; but often, the task of feature construction and selection is tedious and nonscalable. In recent years, however, there have been numerous projects with the goal of constructing general-purpose or domain-specific knowledge bases with entity-relationship-entity triples extracted from various Web sources or collected from user communities, e.g. YAGO, DBpedia, Freebase, UMLS, etc. This paper advocates the simple and yet far-reaching idea that the structured knowledge contained in such knowledge bases can be exploited to automatically extract features for general learning tasks. We introduce an expressive graph-based language for extracting features from such knowledge bases and a theoretical framework for constructing feature vectors from the extracted features. Our experimental evaluation on different learning scenarios provides evidence that the features derived through our framework can considerably improve the prediction accuracy, especially when the labeled data at hand is sparse.

Our work aims at building probabilistic tools for constructing and maintaining large-scale knowledge bases containing entity-relationship-entity triples (statements) extracted from the Web. In order to mitigate the uncertainty inherent in information extraction and integration we propose leveraging the "wisdom of the crowds" by aggregating truth assessments that users provide about statements. The suggested method, CoBayes, operates on a collection of statements, a set of deduction rules (e.g. transitivity), a set of users, and a set of truth assessments of users about statements. We propose a joint probabilistic model of the truth values of statements and the expertise of users for assessing statements. The truth values of statements are interconnected through derivations based on the deduction rules. The correctness of a user's assessment for a given statement is modeled by linear mappings from user descriptions and statement descriptions into a common latent knowledge space where the inner product between user and statement vectors determines the probability that the user assessment for that statement will be correct. Bayesian inference in this complex graphical model is performed using mixed variational and expectation propagation message passing. We demonstrate the viability of CoBayes in comparison to other approaches, on realworld datasets and user feedback collected from Amazon Mechanical Turk.

The proliferation of knowledge-sharing communities and the advances in information extraction have enabled the construction of large knowledge bases using the RDF data model to represent entities and relationships. However, as the Web and its latently embedded facts evolve, a knowledge base can never be complete and up-to-date. On the other hand, a rapidly increasing suite of Web services provide access to timely and high-quality information, but this is encapsulated by the service interface. We propose to leverage the information that could be dynamically obtained from Web services in order to enrich RDF knowledge bases on the fly whenever the knowledge base does not suffice to answer a user query.

Many modern applications are faced with the task of knowledge discovery in entity-relationship graphs, such as domain-specific knowledge bases or social networks. Mining an "informative" subgraph that can explain the relations between k(>= 2) given entities of interest is a frequent knowledge discovery scenario on such graphs. We present MING, a principled method for extracting an informative subgraph for given query nodes. MING builds on a new notion of informativeness of nodes. This is used in a random-walk-with-restarts process to compute the informativeness of entire subgraphs.

Large graphs and networks are abundant in modern information systems: entity-relationship graphs over relational data or Web-extracted entities, biological networks, social online communities, knowledge bases, and many more. Often such data comes with expressive node and edge labels that allow an interpretation as a semantic graph, and edge weights that reflect the strengths of semantic relations between entities. Finding close relationships between a given set of two, three, or more entities is an important building block for many search, ranking, and analysis tasks. From an algorithmic point of view, this translates into computing the best Steiner trees between the given nodes, a classical NP-hard problem. In this paper, we present a new approximation algorithm, coined STAR, for relationship queries over large relationship graphs. We prove that for n query entities, STAR yields an O(log(n))-approximation of the optimal Steiner tree in pseudopolynomial run-time, and show that in practical cases the results returned by STAR are qualitatively comparable to or even better than the results returned by a classical 2-approximation algorithm. We then describe an extension to our algorithm to return the top-k Steiner trees. Finally, we evaluate our algorithm over both main-memory as well as completely diskresident graphs containing millions of nodes. Our experiments show that in terms of efficiency STAR outperforms the best state-of-the-art database methods by a large margin, and also returns qualitatively better results.

We introduce a multi-label classification model and algorithm for labeling heterogeneous networks, where nodes belong to different types and different types have different sets of classification labels. We present a graph-based approach which models the mutual influence between nodes in the network as a random walk. When viewing class labels as "colors", the random surfer is "spraying" different node types with different color palettes; hence the name Graffiti. We demonstrate the performance gains of our method by comparing it to three state-of-the-art techniques for graph-based classification.

We present ANGIE, a system that can answer user queries by combining knowledge from a local database with knowledge retrieved from Web services. If a user poses a query that cannot be answered by the local database alone, ANGIE calls the appropriate Web services to retrieve the missing information. This information is integrated seamlessly and transparently into the local database, so that the user can query and browse the knowledge base while appropriate Web services are called automatically in the background.

The Web has the potential to become the world's largest knowledge base. In order to unleash this potential, the wealth of information available on the Web needs to be extracted and organized. There is a need for new querying techniques that are simple and yet more expressive than those provided by standard keyword-based search engines. Searching for knowledge rather than Web pages needs to consider inherent semantic structures like entities (person, organization, etc.) and relationships (isA, located In, etc.). In this paper, we propose NAGA, a new semantic search engine. NAGA builds on a knowledge base, which is organized as a graph with typed edges, and consists of millions of entities and relationships extracted from Web-based corpora. A graph-based query language enables the formulation of queries with additional semantic information. We introduce a novel scoring model, based on the principles of generative language models, which formalizes several notions such as confidence, informativeness and compactness and uses them to rank query results. We demonstrate NAGA's superior result quality over state-of-the-art search engines and question answering systems.

The presence of encyclopedic Web sources, such as Wikipedia, the Internet Movie \nDatabase (IMDB), World Factbook, etc. calls for new querying techniques that \nare simple and yet more expressive than those provided by standard \nkeyword-based search engines. Searching for explicit knowledge needs to \nconsider inherent semantic structures involving entities and relationships.\n\nIn this demonstration proposal, we describe a semantic search system named \nNAGA. NAGA operates on a knowledge graph, which contains millions of entities \nand relationships derived from various encyclopedic Web sources, such as the \nones above. NAGA's graph-based query language is geared towards expressing \nqueries with additional semantic information. Its scoring model is based on the \nprinciples of generative language models, and formalizes several desiderata \nsuch as confidence, informativeness and compactness of answers.\n\nWe propose a demonstration of NAGA which will allow users to browse the \nknowledge base through a user interface, enter queries in NAGA's query language \nand tune the ranking parameters to test various ranking aspects.

In a recent paper, Martens et al. introduced a specification mechanism for XML tree languages, based on rules of the form (r,s), wherer, s are regular expressions. Sets of such rules can be interpreted in an existential or a universal fashion. An XML tree is existentially valid with respect to a rule set, if for each node there is a rule such that the root path of the node matches r and the children sequence of the node matchess. It is universally valid if each node matching r also matchess. This paper investigates the complexity of reasoning about such rule sets, in particular the satisfiability and the implication problem. Whereas, in general these reasoning problems are complete for EXPTIME, two important fragments are identified with PSPACE and PTIME complexity, respectively.

Current keyword-oriented search engines for theWorld WideWeb do not allow specifying the semantics of queries. We address this limitation with NAGA1, a new semantic search engine. NAGA builds on a large semantic knowledge base of binary relationships (facts) derived from the Web. NAGA provides a simple, yet expressive query language to query this knowledge base. The results are then ranked with an intuitive scoring mechanism. We show the effectiveness and utility of NAGA by comparing its output with that of Googleon some interesting queries.

We present YAGO, a light-weight and extensible ontology with high coverage and quality. YAGO builds on entities and relations and currently contains more than 1 million entities and 5 million facts. This includes the Is-A hierarchy as well as non-taxonomic relations between entities (such as HASONEPRIZE). The facts have been automatically extracted from Wikipedia and unified with WordNet, using a carefully designed combination of rule-based and heuristic methods described in this paper. The resulting knowledge base is a major step beyond WordNet: in quality by adding knowledge about individuals like persons, organizations, products, etc. with their semantic relationships - and in quantity by increasing the number of facts by more than an order of magnitude. Our empirical evaluation of fact correctness shows an accuracy of about 95%. YAGO is based on a logically clean model, which is decidable, extensible, and compatible with RDFS. Finally, we show how YAGO can be further extended by state-of-the-art information extraction techniques.