• Introduces DepAD for using variable dependencies for detecting anomalies. • Reframes unsupervised detection as supervised feature selection and prediction. • Evaluates DepAD against nine state-of-the-art methods on 32 datasets. • DepAD outperforms in detecting anomalies with better interpretability. • Demonstrates new insights and interpretations for detected anomalies. Anomaly detection is crucial for identifying unusual behaviors in data, which often provide valuable insights. This paper introduces Dependency-based Anomaly Detection (DepAD), a general and modular framework that leverages variable dependencies to uncover meaningful anomalies with improved interpretability. DepAD reframes unsupervised anomaly detection as a sequence of supervised feature selection and prediction tasks, enabling users to tailor detection methods to their data and domain. We systematically evaluate 125 DepAD algorithm variants across 32 real-world datasets, combining different off-the-shelf feature selection and prediction techniques. We compare DepAD with twelve state-of-the-art anomaly detection methods and demonstrate its consistent superior performance. Furthermore, we demonstrate that DepAD provides intuitive and informative interpretations of detected anomalies, highlighting its utility in practical applications.

In high-stakes domains such as healthcare, finance, and law, the need for explainable AI is critical. Traditional methods for generating attribution maps, including white-box approaches relying on gradients and black-box techniques that perturb inputs, face challenges like gradient vanishing, blurred attributions, and computational inefficiencies. To overcome these limitations, we introduce a novel approach that leverages diffusion models within the framework of Information Bottleneck (IB) theory. By utilizing the Gaussian noise from diffusion models, we connect the information bottleneck with the Minimum Mean Squared Error (MMSE) from classical information theory, enabling precise calculation of mutual information. This connection leads to a new loss function that minimizes the Signal-to-Noise Ratio (SNR), facilitating efficient optimization and producing high-resolution, pixel-level attribution maps. Our method achieves greater clarity and accuracy in attributions than existing techniques, requiring significantly fewer pixel values to reach the necessary predictive confidence. This work demonstrates the power of diffusion models in advancing explainable AI, particularly in identifying critical input features with high precision.

Zero-inflated data pose significant challenges in precipitation forecasting due to the predominance of zeros with sparse non-zero events. To address this, we propose the Zero Inflation Diffusion Framework (ZIDF), which integrates Gaussian perturbation for smoothing zero-inflated distributions, Transformer-based prediction for capturing temporal patterns, and diffusion-based denoising to restore the original data structure. In our experiments, we use observational precipitation data collected from South Australia along with synthetically generated zero-inflated data. Results show that ZIDF demonstrates significant performance improvements over multiple state-of-the-art precipitation forecasting models, achieving up to 56.7% reduction in MSE and 21.1% reduction in MAE relative to the baseline Non-stationary Transformer. These findings highlight ZIDF’s ability to robustly handle sparse time series data and suggest its potential generalizability to other domains where zero inflation is a key challenge.

Global Climate Models (GCMs) are crucial for predicting future climate changes by simulating the Earth systems. However, GCM outputs exhibit systematic biases due to model uncertainties, parameterization simplifications, and inadequate representation of complex climate phenomena. Traditional bias correction methods, which rely on historical observation data and statistical techniques, often neglect unobserved confounders, leading to biased results. This paper proposes a novel bias correction approach to utilize both GCM and observational data to learn a factor model that captures multi-cause latent confounders. Inspired by recent advances in causality based time series deconfounding, our method first constructs a factor model to learn latent confounders from historical data and then applies them to enhance the bias correction process using advanced time series forecasting models. The experimental results demonstrate significant improvements in the accuracy of precipitation outputs. By addressing unobserved confounders, our approach offers a robust and theoretically grounded solution for climate model bias correction.

Off-policy evaluation (OPE) is a crucial problem in reinforcement learning (RL), where the goal is to estimate the long-term cumulative reward of a target policy using historical data generated by a potentially different behaviour policy. In many real-world applications, such as precision medicine and recommendation systems, unobserved confounders may influence the action, reward, and state transition dynamics, which leads to biased estimates if not properly addressed. While existing methods for handling unobserved confounders in OPE focus on single-action settings, they are less effective in multi-action scenarios commonly found in practical applications, where an agent can take multiple actions simultaneously. In this paper, we propose a novel auxiliary variable-aided method for OPE in multi-action settings with unobserved confounders. Our approach overcomes the limitations of traditional auxiliary variable methods for multi-action scenarios by requiring only a single auxiliary variable, relaxing the need for as many auxiliary variables as the actions. Through theoretical analysis, we prove that our method provides an unbiased estimation of the target policy value. Empirical evaluations demonstrate that our estimator achieves better performance compared to existing baseline methods, highlighting its effectiveness and reliability in addressing unobserved confounders in multi-action OPE settings.

Abstract Over the last few decades, machine learning (ML) applications have grown exponentially, yielding several benefits to society. However, these benefits are tempered with concerns of discriminatory behaviours exhibited by ML models. In this regard, fairness in machine learning has emerged as a priority research area. Consequently, several fairness metrics and algorithms have been developed to mitigate against discriminatory behaviours that ML models may possess. Yet still, very little attention has been paid to the problem of naturally occurring changes in data patterns (aka data distributional drift), and its impact on fairness algorithms and metrics. In this work, we study this problem comprehensively by analyzing 4 fairness-unaware baseline algorithms and 7 fairness-aware algorithms, carefully curated to cover the breadth of its typology, across 5 datasets including public and proprietary data, and evaluated them using 3 predictive performance and 10 fairness metrics. In doing so, we show that (1) data distributional drift is not a trivial occurrence, and in several cases can lead to serious deterioration of fairness in so-called fair models; (2) contrary to some existing literature, the size and direction of data distributional drift is not correlated to the resulting size and direction of unfairness; and (3) choice of, and training of fairness algorithms is impacted by the effect of data distributional drift which is largely ignored in the literature. Emanating from our findings, we synthesize several policy implications of data distributional drift on fairness algorithms that can be very relevant to stakeholders and practitioners.

Latent confounders are a fundamental challenge for inferring causal effects from observational data. The instrumental variable (IV) approach is a practical way to address this challenge. Existing IV-based estimators need a known IV or other strong assumptions, such as the existence of two or more IVs in the system, which limits the application of the IV approach. In this article, we consider a relaxed requirement, which assumes there is an IV proxy in the system without knowing which variable is the proxy. We propose a variational autoencoder (VAE)-based disentangled representation learning method to learn an IV representation from a dataset with latent confounders and then utilize the IV representation to obtain an unbiased estimation of the causal effect from the data. Extensive experiments on synthetic and real-world data have demonstrated that the proposed algorithm outperforms the existing IV-based estimators and VAE-based estimators.

Causal inference from longitudinal observational data is a challenging problem due to the difficulty in correctly identifying the time-dependent confounders, especially in the presence of latent time-dependent confounders. Instrumental variable (IV) is a powerful tool for addressing the latent confounders issue, but the traditional IV technique cannot deal with latent time-dependent confounders in longitudinal studies. In this work, we propose a novel Time-dependent Instrumental Factor Model (TIFM) for time-varying causal effect estimation from data with latent time-dependent confounders. At each time-step, the proposed TIFM method employs the Recurrent Neural Network (RNN) architecture to infer latent IV, and then uses the inferred latent IV factor for addressing the confounding bias caused by the latent time-dependent confounders. We provide a theoretical analysis for the proposed TIFM method regarding causal effect estimation in longitudinal data. Extensive evaluation with synthetic datasets demonstrates the effectiveness of TIFM in addressing causal effect estimation over time. We further apply TIFM to a climate dataset to showcase the potential of the proposed method in tackling real-world problems.

In the pursuit of intersectional group fairness in machine learning models, significant attention has been directed towards fair representation learning methods. These methods aim to mitigate bias in training data by encoding data effectively while removing sensitive attribute information. However, existing fair representation learning methods often assume that decoupling sensitive attribute information from the latent representation will automatically lead to fairness on any downstream tasks learnt on the non-sensitive subspace of the latent representation. Nonetheless, biases can persist even when using representations devoid of sensitive attribute information. This is due to the learning algorithm's influence during downstream task training. In this paper, we propose a method dubbed FairReg which integrates fairness regularization with fair representation learning. This unified approach creates a more comprehensive and robust framework for ensuring intersectional group fairness in machine learning models. Empirical evaluations conducted on two real-world depression prediction datasets demonstrate the effectiveness of our method in improving intersectional group fairness compared to existing approaches.

This paper studies the challenging problem of estimating causal effects from observational data, in the presence of unobserved confounders. The two-stage least square (TSLS) method and its variants with a standard instrumental variable (IV) are commonly used to eliminate confounding bias, including the bias caused by unobserved confounders, but they rely on the linearity assumption. Besides, the strict condition of unconfounded instruments posed on a standard IV is too strong to be practical. To address these challenging and practical problems of the standard IV method (linearity assumption and the strict condition), in this paper, we use a conditional IV (CIV) to relax the unconfounded instrument condition of standard IV and propose a non-linear \underline{CIV} regression with \underline{C}onfounding \underline{B}alancing \underline{R}epresentation \underline{L}earning, CBRL.CIV, for jointly eliminating the confounding bias from unobserved confounders and balancing the observed confounders, without the linearity assumption. We theoretically demonstrate the soundness of CBRL.CIV. Extensive experiments on synthetic and two real-world datasets show the competitive performance of CBRL.CIV against state-of-the-art IV-based estimators and superiority in dealing with the non-linear situation.

An essential and challenging problem in causal inference is causal effect estimation from observational data. The problem becomes more difficult with the presence of unobserved confounding variables. The front-door adjustment is an approach for dealing with unobserved confounding variables. However, the restriction for the standard front-door adjustment is difficult to satisfy in practice. In this paper, we relax some of the restrictions by proposing the concept of conditional front-door (CFD) adjustment and develop the theorem that guarantees the causal effect identifiability of CFD adjustment. By leveraging the ability of deep generative models, we propose CFDiVAE to learn the representation of the CFD adjustment variable directly from data with the identifiable Variational AutoEncoder and formally prove the model identifiability. Extensive experiments on synthetic datasets validate the effectiveness of CFDiVAE and its superiority over existing methods. The experiments also show that the performance of CFDiVAE is less sensitive to the causal strength of unobserved confounding variables. We further apply CFDiVAE to a real-world dataset to demonstrate its potential application.

Instrumental variable (IV) is a powerful approach to inferring the causal effect of a treatment on an outcome of interest from observational data even when there exist latent confounders between the treatment and the outcome. However, existing IV methods require that an IV is selected and justified with domain knowledge. An invalid IV may lead to biased estimates. Hence, discovering a valid IV is critical to the applications of IV methods. In this article, we study and design a data-driven algorithm to discover valid IVs from data under mild assumptions. We develop the theory based on partial ancestral graphs (PAGs) to support the search for a set of candidate ancestral IVs (AIVs), and for each possible AIV, the identification of its conditioning set. Based on the theory, we propose a data-driven algorithm to discover a pair of IVs from data. The experiments on synthetic and real-world datasets show that the developed IV discovery algorithm estimates accurate estimates of causal effects in comparison with the state-of-the-art IV-based causal effect estimators.

The instrumental variable (IV) approach is a widely used way to estimate the causal effects of a treatment on an outcome of interest from observational data with latent confounders. A standard IV is expected to be related to the treatment variable and independent of all other variables in the system. However, it is challenging to search for a standard IV from data directly due to the strict conditions. The conditional IV (CIV) method has been proposed to allow a variable to be an instrument conditioning on a set of variables, allowing a wider choice of possible IVs and enabling broader practical applications of the IV approach. Nevertheless, there is not a data-driven method to discover a CIV and its conditioning set directly from data. To fill this gap, in this paper, we propose to learn the representations of the information of a CIV and its conditioning set from data with latent confounders for average causal effect estimation. By taking advantage of deep generative models, we develop a novel data-driven approach for simultaneously learning the representation of a CIV from measured variables and generating the representation of its conditioning set given measured variables. Extensive experiments on synthetic and real-world datasets show that our method outperforms the existing IV methods.

Estimating direct and indirect causal effects from observational data is crucial to understanding the causal mechanisms and predicting the behaviour under different interventions. Causal mediation analysis is a method that is often used to reveal direct and indirect effects. Deep learning shows promise in mediation analysis, but the current methods only assume latent confounders that affect treatment, mediator and outcome simultaneously, and fail to identify different types of latent confounders (e.g., confounders that only affect the mediator or outcome). Furthermore, current methods are based on the sequential ignorability assumption, which is not feasible for dealing with multiple types of latent confounders. This work aims to circumvent the sequential ignorability assumption and applies the piecemeal deconfounding assumption as an alternative. We propose the Disentangled Mediation Analysis Variational AutoEncoder (DMAVAE), which disentangles the representations of latent confounders into three types to accurately estimate the natural direct effect, natural indirect effect and total effect. Experimental results show that the proposed method outperforms existing methods and has strong generalisation ability. We further apply the method to a real-world dataset to show its potential application.

Researchers have developed various visualization techniques, such as attribution maps, to understand which parts of an input contribute most to a model’s decision. However, existing methods often produce disparate results and may lack human-perceptual interpretability. In this work, we propose Relevance-IBA, a novel approach that combines the strengths of Information Bottleneck Attribution (IBA) and Layer-wise Relevance Propagation’s (LRP) method to estimate more accurate and human-perceptually interpretable attribution maps. Our method accentuates the contours and subtle details of the identified object, making the model’s decisions more intuitively understandable. Additionally, we introduce a segmentation-oriented evaluation technique, which assesses the capacity of interpretability methods by emphasizing the most important pixels within an object’s boundaries. We benchmark Relevance-IBA against various methodologies, including DeepLIFT, Integrated Gradients, Guided-BP, Guided-GradCAM, IBA, and InputIBA. Our results indicate that Relevance-IBA not only boosts attribution accuracy but also prioritizes human-perceptual clarity, making it a valuable tool for interpreting complex model behaviors.

In causal inference, it is a fundamental task to estimate the causal effect from observational data. However, latent confounders pose major challenges in causal inference in observational data, for example, confounding bias and M-bias. Recent data-driven causal effect estimators tackle the confounding bias problem via balanced representation learning, but assume no M-bias in the system, thus they fail to handle the M-bias. In this paper, we identify a challenging and unsolved problem caused by a variable that leads to confounding bias and M-bias simultaneously. To address this problem with co-occurring M-bias and confounding bias, we propose a novel Disentangled Latent Representation learning framework for learning latent representations from proxy variables for unbiased Causal effect Estimation (DLRCE) from observational data. Specifically, DLRCE learns three sets of latent representations from the measured proxy variables to adjust for the confounding bias and M-bias. Extensive experiments on both synthetic and three real-world datasets demonstrate that DLRCE significantly outperforms the state-of-the-art estimators in the case of the presence of both confounding bias and M-bias.

In Disability Employment Services (DES), an emerging problem is to make job survival analysis stable in unseen environments without prior knowledge of these environments. Existing survival analysis methods cannot adequately solve this problem since they assume that distribution of unseen data is similar to that observed during training. However, this assumption can be violated in practice where unanticipated events such as COVID19 and inflation can change the work and life patterns of people with disability. Models trained before the COVID19 pandemic may make unreliable job survival predictions in COVID19 or inflation situations. It is also costly and time consuming to frequently re-train and deploy the models. This paper proposes a stable survival analysis method for the DES sector without requiring prior knowledge of deployment environments. Latent representations are learned to capture non-linear relationships between relevant features and job survival time. Two reweighting stages are developed to remove censoring and conditional spurious correlations between irrelevant features and the survival outcome. The case study of Australian workers with disability shows that our method can make stable risk predictions. It can also help workers with disability determine the most effective skills for improvement to increase their job survival time. Further evaluations with public datasets show the promising stable performance of our method in other applications.

In recent years, predictive models have been increasingly used by education practitioners and stakeholders to leverage actionable insights to support student success. Usually, model selection (i.e., the decision of which predictive model to use) is largely based on the predictive performance of the models. Nevertheless, it has become important to consider fairness as an integral part of the criteria for model selection. Might a model be unfair towards certain demographic groups? Might it systematically perform poorly for certain demographic groups? Indeed, prior studies affirm this. Which model out of the lot should we choose then? Additionally, prior studies suggest demographic group imbalance in the training dataset to be a source of such unfairness. If so, would the fairness of the predictive models improve if the demographic group distribution in the training dataset becomes balanced? This study seeks to answer these questions. Firstly, we analyze the fairness of 4 of the commonly used state-of-the-art models to predict course success for 3 IT courses in a large public Australian university. Specifically, we investigate if the models serve different demographic groups equally. Secondly, to address the identified unfairnes---supposedly caused by the demographic group imbalance---we train the models on 3 types ofbalanced data and investigate again if the unfairness was mitigated. We found that none of the predictive models wasconsistently fair in all 3 courses. This suggests that model selection decision should be carefully made by both the researchers and stakeholders as the per the requirement of the domain of application. Furthermore, we found that balancing demographic groups (and class labels) is not enough---albeit can be an initial step---to ensure fairness of predictive models in education. An implication of this is that sometimes, the source of unfairness may not be immediately apparent. Therefore, "blindly" attributing the unfairness to demographic group imbalance may cause the unfairness to persist even when the data becomes balanced. We hope that our findings can guide practitioners and relevant stakeholders in making well-informed decisions.

Abstract In personalised decision making, evidence is required to determine whether an action (treatment) is suitable for an individual. Such evidence can be obtained by modelling treatment effect heterogeneity in subgroups. The existing interpretable modelling methods take a top-down approach to search for subgroups with heterogeneous treatment effects and they may miss the most specific and relevant context for an individual. In this paper, we design a Treatment effect pattern (TEP) to represent treatment effect heterogeneity in data. To achieve an interpretable presentation of TEPs, we use a local causal structure around the outcome to explicitly show how those important variables are used in modelling. We also derive a formula for unbiasedly estimating the Conditional Average Causal Effect (CATE) using the local structure in our problem setting. In the discovery process, we aim at minimising heterogeneity within each subgroup represented by a pattern. We propose a bottom-up search algorithm to discover the most specific patterns fitting individual circumstances the best for personalised decision making. Experiments show that the proposed method models treatment effect heterogeneity better than three other existing tree based methods in synthetic and real world data sets.

An emerging problem in Disability Employment Services (DES) is recommending to people with disability the right skill to upgrade and the right upgrade level to achieve maximum improvement in their employment success. This problem requires causal reasoning to estimate the individual causal effect of possible factors on the outcome to determine the most effective intervention. In this paper, we propose a causal graph based framework to solve the intervention recommendation problem for survival outcome (job retention time) and non-survival outcome (employment status). For an individual, a personalized causal graph is predicted for them. It indicates which factors affect the outcome and their causal effects at different intervention levels. Based on the causal graph, we can determine the most effective intervention for an individual, i.e., the one that can generate a maximum outcome increase. Experiments with two case studies show that our framework can help people with disability increase their employment success. Evaluations with public datasets also show the advantage of our framework in other applications.

Causal effect estimation from observational data is a challenging problem, especially with high dimensional data and in the presence of unobserved variables. The available data-driven methods for tackling the problem either provide an estimation of the bounds of a causal effect (i.e., nonunique estimation) or have low efficiency. The major hurdle for achieving high efficiency while trying to obtain unique and unbiased causal effect estimation is how to find a proper adjustment set for confounding control in a fast way, given the huge covariate space and considering unobserved variables. In this paper, we approach the problem as a local search task for finding valid adjustment sets in data. We establish the theorems to support the local search for adjustment sets, and we show that unique and unbiased estimation can be achieved from observational data even when there exist unobserved variables. We then propose a data-driven algorithm that is fast and consistent under mild assumptions. We also make use of a frequent pattern mining method to further speed up the search of minimal adjustment sets for causal effect estimation. Experiments conducted on extensive synthetic and real-world datasets demonstrate that the proposed algorithm outperforms the state-of-the-art criteria/estimators in both accuracy and time-efficiency.

Causal effect estimation from observational data is a crucial but challenging task. Currently, only a limited number of data-driven causal effect estimation methods are available. These methods either provide only a bound estimation of causal effects of treatment on the outcome or generate a unique estimation of the causal effect but making strong assumptions on data and having low efficiency. In this article, we identify a problem setting with the Cause Or Spouse of the treatment Only (COSO) variable assumption and propose an approach to achieving a unique and unbiased estimation of causal effects from data with hidden variables. For the approach, we have developed the theorems to support the discovery of the proper covariate sets for confounding adjustment (adjustment sets). Based on the theorems, two algorithms are proposed for finding the proper adjustment sets from data with hidden variables to obtain unbiased and unique causal effect estimation. Experiments with synthetic datasets generated using five benchmark Bayesian networks and four real-world datasets have demonstrated the efficiency and effectiveness of the proposed algorithms, indicating the practicability of the identified problem setting and the potential of the proposed approach in real-world applications.

The ultimate goal in adversarial defense is to build a universally robust defense against all types of attacks, but ongoing arms race between adversarial attacks and defenses show the difficulty in building a deterministic defense to work towards the goal. Leveraging the idea of a mixture model, in this paper, we introduce a new Randomized Adversarial Defense method (RAD) to increase the robustness against adversarial examples. RAD is designed as a simple, yet effective random mixture of a global model and one or more local models. The mixture is able to create a random decision boundary for a test instance, making it harder for an adversarial example to succeed, and thus increasing the robustness of the defense. The global model is adversarially trained to provide the baseline robustness. The local models are aimed to supplement the global model and thus the decision boundary of a local model is expected to be adjacent to the decision boundary of the global model. These models then together form a random mixture to create a randomized (non-deterministic) decision boundary for each test instance at the end. Such a randomization scheme reduces the adversarial risk since the adversary has to approximate the best attack despite of the given complete knowledge of the parameters of the individual classifiers. By proposing the notion of having a global and local models with different focuses in the mixture and the way of creating a local model which has minimum dependency on the base (global) model, RAD provides a simpler and more flexible, yet effective approach building a randomized defense, compared with the existing randomization based methods. Experimental results show that our simple randomization approach outperforms the most robust deterministic defense method and performs competitively upon the existing randomized defense method against strong adaptive attacks on CIFAR10 and CIFAR100.

In Disability Employment Service (DES), DES providers are confronted with "what-if" questions to assist workers with disability in deciding which skill should be improved to increase their job retention time. For instance, what would happen to the job retention time of a worker with disability if he improved his computer skill to an advanced level? This requires counterfactual inference to estimate the counterfactuals of the survival outcome, i.e., job retention time, under different skill improvement scenarios. While exiting survival analysis techniques are not designed for counterfactual problems, current counterfactual prediction methods are assumed to work with non-survival outcomes. In this paper, we propose the Counterfactual Survival Network (CSN), a representation learning based method for counterfactual survival prediction, where both confounding and censoring biases are removed based on latent representations. Since ground truth counterfactuals are unavailable, we develop a sample specific estimator to estimate counterfactuals for training samples. These estimated counterfactual outcomes are used as pseudo ground truth to train the counterfactual prediction model. We demonstrate the benefits of our method in decision support tasks with the case study of Australian workers and three public datasets. Results show that CSN can help Australian workers with disability increase their job retention time. Our method also shows its promising performance in other applications.

Unobserved confounding is the main obstacle to causal effect estimation from observational data. Instrumental variables (IVs) are widely used for causal effect estimation when there exist latent confounders. With the standard IV method, when a given IV is valid, unbiased estimation can be obtained, but the validity requirement on a standard IV is strict and untestable. Conditional IVs have been proposed to relax the requirement of standard IVs by conditioning on a set of observed variables (known as a conditioning set for a conditional IV). However, the criterion for finding a conditioning set for a conditional IV needs a directed acyclic graph (DAG) representing the causal relationships of both observed and unobserved variables. This makes it challenging to discover a conditioning set directly from data. In this paper, by leveraging maximal ancestral graphs (MAGs) for causal inference with latent variables, we study the graphical properties of ancestral IVs, a type of conditional IVs using MAGs, and develop the theory to support data-driven discovery of the conditioning set for a given ancestral IV in data under the pretreatment variable assumption. Based on the theory, we develop an algorithm for unbiased causal effect estimation with a given ancestral IV and observational data. Extensive experiments on synthetic and real-world datasets demonstrate the performance of the algorithm in comparison with existing IV methods.

In Disability Employment Services (DES), an emerging problem is recommending to disabled workers the right skill to upgrade and the right upgrade level to achieve a maximum increase in their job retention time. This problem involves causal reasoning to estimate the individual causal effect (ICE) on the survival outcome, i.e., job retention time, to determine the most effective intervention for a worker. Existing methods are not suitable to solve our problem. They are mostly developed for non-causal or non-survival challenges, while methods for causal survival analysis are under-explored. This paper proposes a representation learning method for recommending personalized interventions that can generate a maximum increase in job retention time for workers with disability. In our method, observed covariates are disentangled into latent variables based on which confounding and censoring biases are eliminated, and the ICE prediction model is built. Since true ICE values are not directly measurable in observational data, a reverse engineering technique is developed to estimate ICE for training samples. These estimated ICE values are then used as the pseudo ground truth to train the prediction model. Experiments with a case study of Australian workers with disability show that by adopting personalized interventions recommended by our method, disabled workers can increase their job retention time by up to 2.8 months. Additional evaluations with public datasets also show the technical strengths of our method in other applications.

Abstract A large number of covariates can have a negative impact on the quality of causal effect estimation since confounding adjustment becomes unreliable when the number of covariates is large relative to the number of samples. Propensity score is a common way to deal with a large covariate set, but the accuracy of propensity score estimation (normally done by logistic regression) is also challenged by the large number of covariates. In this paper, we prove that a large covariate set can be reduced to a lower dimensional representation which captures the complete information for adjustment in causal effect estimation. The theoretical result enables effective data-driven algorithms for causal effect estimation. Supported by the result, we develop an algorithm that employs a supervised kernel dimension reduction method to learn a lower dimensional representation from the original covariate space, and then utilises nearest neighbour matching in the reduced covariate space to impute the counterfactual outcomes to avoid the large sized covariate set problem. The proposed algorithm is evaluated on two semisynthetic and three real-world datasets and the results show the effectiveness of the proposed algorithm.

Despite their high accuracy, deep neural networks (DNNs) are vulnerable to adversarial examples. Currently, adversarial training is the mainstream defense approach against adversarial examples. However, given the unknown nature of adversarial attacks in real life, this approach has fundamental limitations in practical use, as it is impossible to obtain sufficient adversarial examples for the training. In this paper, we propose RanTrain, a simple training approach which employs a background class with random noise images to augment the original DNN model and training data, without requiring any adversarial examples. Experiments have shown that RanTrain works effectively with different datasets and various DNN structures, and it significantly increases the robustness of DNNs to adversarial examples.

In Disability Employment Services (DES), a growing problem is recommending to disabled job seekers which skill should be upgraded and the best level for upgrading this skill to increase their employment potential most. This problem involves counterfactual reasoning to infer causal effect of factors on employment status to recommend the most effective intervention. Related methods cannot solve our problem adequately since they are developed for non-counterfactual challenges, for binary causal factors, or for randomized trials. In this paper, we present a causality-based method to tackle the problem. The method includes two stages where causal factors of employment status are first detected from data. We then combine a counterfactual reasoning framework with a machine learning approach to build an interpretable model for generating personalized recommendations. Experiments on both synthetic datasets and a real case study from a DES provider show consistent promising performance of improving employability of disabled job seekers. Results from the case study disclose effective factors and their best levels for intervention to increase employability. The most effective intervention varies among job seekers. Our model can separate job seekers by degree of employability increase. This is helpful for DES providers to allocate resources for employment assistance. Moreover, causal interpretability makes our recommendations actionable in DES business practice.

In the disability employment service, an emerging challenge is to recommend the skills whose improvement will increase the employment perspective most. The process of a skill being improved is called an intervention and different skills are called factors. The problem involves recommendation for outcome improvement, which requires estimating the improvement in the employment perspective, i.e., the outcome, driven by interventions on recommended factors. Currently, most recommendation systems deployed for the employment service rely on traditional recommendation models where the desired outcome instead of the degree of outcome improvement is the main goal for optimization. In this paper, we present a causality-based approach for recommending factors for intervention to achieve the largest improvement in the employment potential of disabled job seekers. It involves inferring the causal effect of interventions on the employment outcome to make recommendations for individuals. The causal interpretation of our model can justify given recommendations. We conduct a case study with our industry partner in the disability employment service. Results show that the recommended interventions could improve the employability of disabled job seekers. Experiments are also carried out with datasets in other domains to demonstrate the promise of our approach in different applications.

This paper discusses the problem of causal query in observational data with hidden variables, with the aim of seeking the change of an outcome when "manipulating" a variable while given a set of plausible confounding variables which affect the manipulated variable and the outcome. Such an "experiment on data" to estimate the causal effect of the manipulated variable is useful for validating an experiment design using historical data or for exploring confounders when studying a new relationship. However, existing data-driven methods for causal effect estimation face some major challenges, including poor scalability with high dimensional data, low estimation accuracy due to heuristics used by the global causal structure learning algorithms, and the assumption of causal sufficiency when hidden variables are inevitable in data. In this paper, we develop theorems for using local search to find a superset of the adjustment (or confounding) variables for causal effect estimation from observational data under a realistic pretreatment assumption. The theorems ensure that the unbiased estimate of causal effect is included in the set of causal effects estimated by the superset of adjustment variables. Based on the developed theorems, we propose a data-driven algorithm for causal query. Experiments show that the proposed algorithm is faster and produces better causal effect estimation than an existing data-driven causal effect estimation method with hidden variables. The causal effects estimated by the proposed algorithm are as accurate as those by the state-of-the-art methods using domain knowledge.

Text documents are often mapped to vectors of binary values where 1 indicates the presence of a word and 0 indicates the absence. The vectors are then used to train predictive models. In tree-based ensemble models, predictions from some decision trees may be made purely from absent words. This type of predictions should be trusted less as absent words can be interpreted in multiple ways. In this work, we propose to improve the comprehensibility and accuracy of ensemble models by distinguishing word presence and absence. The presented method weights predictions based on word presence. Experimental results on 35 real text datasets indicate that our method outperforms state-of-the-art ensemble methods on various text classification tasks.

The Random Forest (RF) classifiers are suitable for dealing with the high dimensional noisy data in text classification. An RF model comprises a set of decision trees each of which is trained using random subsets of features. Given an instance, the prediction by the RF is obtained via majority voting of the predictions of all the trees in the forest. However, different test instances would have different values for the features used in the trees and the trees should contribute differently to the predictions. This diverse contribution of the trees is not considered in traditional RFs. Many approaches have been proposed to model the diverse contributions by selecting a subset of trees for each instance. This paper is among these approaches. It proposes a Semantics Aware Random Forest (SARF) classifier. SARF extracts the features used by trees to generate the predictions and selects a subset of the predictions for which the features are relevant to the predicted classes. We evaluated SARF's classification performance on $30$ real-world text datasets and assessed its competitiveness with state-of-the-art ensemble selection methods. The results demonstrate the superior performance of the proposed approach in textual information retrieval and initiate a new direction of research to utilise interpretability of classifiers.

Summary Data publishing is pivotal to advances in knowledge discovery. Nonetheless, such publishing may suffer from privacy disclosures. This is especially the case in transactional data such as web search and point of sales logs. The reason is that the current potent privacy preserving mechanisms mainly focus on relational data. In this work, we propose a new privacy metric for transactional data to prevent inference attacks by ensuring that the adversary learns no more about an intended victim than what is publicly available. We then propose a publication mechanism Anony, which satisfies our privacy metric without excessive loss of utility. Finally, we present an empirical evaluation of our method on three benchmark datasets, and the results show the effectiveness of our method.

Summary Graph is one of the most important data structures to model social networks and becomes popular to find interesting relationships between individuals. Since graphs may contain sensitive information, data curators usually need to anonymize the graph before publication to prevent individual re‐identification, which thus leads to plenty of anonymized graphs for data sharing and exploration. However, the new structures and properties of anonymized graphs make the traditional graph indexing method inefficient or even invalid for query processing. To address the subgraph query problem over anonymized graph database, in this paper, we first introduce basic concepts about anonymized graphs and subgraph queries, then propose an index structure named Closure+‐tree to process the subgraph query efficiently. In particular, graphs were organized hierarchically that each node is an union of its child nodes under some specified mapping functions. During the processing of subgraph queries, the whole graph descendants will be pruned if their union does not contain the query graph. To evaluate the performance of our proposed Closure+‐tree, extensive experiments are performed on both real and synthetic graph data sets. The experimental results revealed that our index structure can prune up to 80% unqualified graphs with variable size of queries. Furthermore, the size of our index structure is only around a quarter of the entire anonymized graph data set, which indicates good scalability over large data sets.

Entity resolution (ER) is the problem of accurately identifying multiple, differing, and possibly contradicting representations of unique real-world entities in data. It is a challenging and fundamental task in data cleansing and data integration. In this work, we propose graph differential dependencies (GDDs) as an extension of the recently developed graph entity dependencies (which are formal constraints for graph data) to enable approximate matching of values. Furthermore, we investigate a special discovery of GDDs for ER by designing an algorithm for generating a non-redundant set of GDDs in labelled data. Then, we develop an effective ER technique, Certus, that employs the learned GDDs for improving the accuracy of ER results. We perform extensive empirical evaluation of our proposals on five real-world ER benchmark datasets and a proprietary database to test their effectiveness and efficiency. The results from the experiments show the discovery algorithm and Certus are efficient; and more importantly, GDDs significantly improve the precision of ER without considerable trade-off of recall.

With more and more decisions being made by learnt algorithms from data, algorithmic discriminations have become a risk for civil rights. The detection of discrimination is a process of counterfactual reasoning. This paper proposes a general detection framework by combining a data mining method with a well established counterfactual reasoning framework, potential outcome model. The potential outcome model supports operational definitions of global and local discriminations and discriminations by combined factors, while a data mining method makes the detection efficient. The proposed method, instantiated by association rule mining with potential outcome model based causal effect estimation, is evaluated with four real world data sets and is compared with a Bayesian network (BN) based detection method. It is able to detect not only global discriminations that are detected by the BN based method, but also local and combined discriminations that the BN based method cannot find. The proposed method is efficient, and scales well with the data set size and the number of attributes.

Summary Voice over Internet Protocol (VoIP) is increasingly used by both individuals and corporate users. Recent studies had suggested that communications using VoIP apps may not be encrypted, although it has not been demonstrated that original conversations could be recovered from unencrypted communications that were subsequently intercepted. In this research, we propose a process for codec identification and decoding of captured communications from 15 popular Android VoIP apps. We are able to recover the original voice conversations from intercepted app‐to‐app voice calls of nine apps and app‐to‐phone calls from 10 apps. Copyright © 2016 John Wiley & Sons, Ltd.

Uncovering causal relationships in data is a major objective of data analytics. Currently, there is a need for scalable and automated methods for causal relationship exploration in data. Classification methods are fast and they could be practical substitutes for finding causal signals in data. However, classification methods are not designed for causal discovery and a classification method may find false causal signals and miss the true ones. In this paper, we develop a causal decision tree (CDT) where nodes have causal interpretations. Our method follows a well-established causal inference framework and makes use of a classic statistical test to establish the causal relationship between a predictor variable and the outcome variable. At the same time, by taking the advantages of normal decision trees, a CDT provides a compact graphical representation of the causal relationships, and the construction of a CDT is fast as a result of the divide and conquer strategy employed, making CDTs practical for representing and finding causal signals in large data sets. Experiment results demonstrate that CDTs can identify meaningful causal relationships and the CDT algorithm is scalable.

Randomised controlled trials (RCTs) are the most effective approach to causal discovery, but in many circumstances it is impossible to conduct RCTs. Therefore, observational studies based on passively observed data are widely accepted as an alternative to RCTs. However, in observational studies, prior knowledge is required to generate the hypotheses about the cause-effect relationships to be tested, and hence they can only be applied to problems with available domain knowledge and a handful of variables. In practice, many datasets are of high dimensionality, which leaves observational studies out of the opportunities for causal discovery from such a wealth of data sources. In another direction, many efficient data mining methods have been developed to identify associations among variables in large datasets. The problem is that causal relationships imply associations, but the reverse is not always true. However, we can see the synergy between the two paradigms here. Specifically, association rule mining can be used to deal with the high-dimensionality problem, whereas observational studies can be utilised to eliminate noncausal associations. In this article, we propose the concept of causal rules (CRs) and develop an algorithm for mining CRs in large datasets. We use the idea of retrospective cohort studies to detect CRs based on the results of association rule mining. Experiments with both synthetic and real-world datasets have demonstrated the effectiveness and efficiency of CR mining. In comparison with the commonly used causal discovery methods, the proposed approach generally is faster and has better or competitive performance in finding correct or sensible causes. It is also capable of finding a cause consisting of multiple variables—a feature that other causal discovery methods do not possess.

To address the frequently occurring situation where data is inexact or imprecise, a number of extensions to the classical notion of a functional dependency (FD) integrity constraint have been proposed in recent years. One of these extensions is the notion of a differential dependency (DD), introduced in the recent article “Differential Dependencies: Reasoning and Discovery” by Song and Chen in the March 2011 edition of this journal. A DD generalises the notion of an FD by requiring only that the values of the attribute from the RHS of the DD satisfy a distance constraint whenever the values of attributes from the LHS of the DD satisfy a distance constraint. In contrast, an FD requires that the values from the attributes in the RHS of an FD be equal whenever the values of the attributes from the LHS of the FD are equal. The article “Differential Dependencies: Reasoning and Discovery” investigated a number of aspects of DDs, the most important of which, since they form the basis for the other topics investigated, were the consistency problem (determining whether there exists a relation instance that satisfies a set of DDs) and the implication problem (determining whether a set of DDs logically implies another DD). Concerning these problems, a number of results were claimed in “Differential Dependencies: Reasoning and Discovery”. In this article we conduct a detailed analysis of the correctness of these results. The outcomes of our analysis are that, for almost every claimed result, we show there are either fundamental errors in the proof or the result is false. For some of the claimed results we are able to provide corrected proofs, but for other results their correctness remains open.

Conditional functional dependencies (CFDs) have been proposed as a new type of semantic rules extended from traditional functional dependencies. They have shown great potential for detecting and repairing inconsistent data. Constant CFDs are 100% confidence association rules. The theoretical search space for the minimal set of CFDs is the set of minimal generators and their closures in data. This search space has been used in the currently most efficient constant CFD discovery algorithm. In this paper, we propose pruning criteria to further prune the theoretic search space, and design a fast algorithm for constant CFD discovery. We evaluate the proposed algorithm on a number of media to large real-world data sets. The proposed algorithm is faster than the currently most efficient constant CFD discovery algorithm, and has linear time performance in the size of a data set.

Functional and inclusion dependency discovery is important to knowledge discovery, database semantics analysis, database design, and data quality assessment. Motivated by the importance of dependency discovery, this paper reviews the methods for functional dependency, conditional functional dependency, approximate functional dependency, and inclusion dependency discovery in relational databases and a method for discovering XML functional dependencies.

Data anonymization has become a major technique in privacy preserving data publishing. Many methods have been proposed to anonymize one dataset and a series of datasets of a data owner. However, no method has been proposed for the anonymization of data of multiple independent data publications. A data owner publishes a dataset, which contains overlapping population with other datasets published by other independent data owners. In this paper we analyze the privacy risk in the such scenario and vulnerability of partitioned based anonymization methods. We show that no partitioned based anonymization methods can protect privacy in arbitrary data distributions, and identify a case that the privacy can be protected in the scenario. We propose a new generalization principle ε-cloning to protect privacy for multiple independent data publications. We also develop an effective algorithm to achieve the ε-cloning. We experimentally show that the proposed algorithm anonymizes data to satisfy the privacy requirement and preserves good data utility.

Using reputation system to help determine trustworthiness of online entities is a widely adopted method in the current web environment. Reputation bootstrapping, known as determining a default reputation value for a newcomer, is an important issue which currently is short of a proper solution. In this paper, we introduce a new concept, marginal reputation utility, based on which we discuss the relationship between trust and reputation and construct a reputation bootstrapping framework that is universally applicable to different reputation models and systems. We use experiments to test the effectiveness of the proposed framework upon different reputation models, which shows satisfactory results.

In this article, we address the problem of how to extend the definition of functional dependencies (FDs) in incomplete relations to XML documents (called XFDs) using the well-known strong satisfaction approach.We propose a syntactic definition of strong XFD satisfaction in an XML document and then justify it by showing that, similar to the case in relational databases, for the case of simple paths, keys in XML are a special case of XFDs. We also propose a normal form for XML documents based on our definition of XFDs and provide a formal justification for it by proving that it is a necessary and sufficient condition for the elimination of redundancy in an XML document.

XML has been accepted as a universal format for data interchange and publication. It can be applied in the applications in which the data of a database needs to be viewed in XML format so that the data being viewed takes more semantics and is easily understood. In these applications, the user of the data to be viewed sees only XML data, not the database. He may use XML query languages such as XSLT to query data and the retrieved data is presented in XML format to them. We are interested in the connection between the data that the user sees and the data in the database. More specifically, we are interested in translating XSLT queries to SQL queries.

Article Maintaining views in object-relational databases Share on Authors: Jixue Liu School of Computer Science, University. of South Australia, Adelaide, SA5084, Australia School of Computer Science, University. of South Australia, Adelaide, SA5084, AustraliaView Profile , Millist Vincent School of Computer Science, University. of South Australia, Adelaide, SA5084, Australia School of Computer Science, University. of South Australia, Adelaide, SA5084, AustraliaView Profile , Mukesh Mohania Dept. of Computer Science, Western Michigan University, Kalamazoo, MI Dept. of Computer Science, Western Michigan University, Kalamazoo, MIView Profile Authors Info & Claims CIKM '00: Proceedings of the ninth international conference on Information and knowledge managementNovember 2000 Pages 102–109https://doi.org/10.1145/354756.354807Online:06 November 2000Publication History 6citation482DownloadsMetricsTotal Citations6Total Downloads482Last 12 Months5Last 6 weeks0 Get Citation AlertsNew Citation Alert added!This alert has been successfully added and will be sent to:You will be notified whenever a record that you have chosen has been cited.To manage your alert preferences, click on the button below.Manage my AlertsNew Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteGet Access

Incremental view maintenance is widely preferred to view recomputation when an update to a base relation is small in size. The update size becomes an important concept for measuring the cheap performance of the incremental maintenance. In this paper, we investigate what the limit of the update size is for incremental maintenance. When the size of an update exceeds the limit, the incremental maintenance is no longer cheaper than the view recomputation. The investigation is based on incremental equations for operators in the nested relational model. We implement these equations in the Informix Universal Database Server. We propose a cost model for the performance analysis of the equations, we analyze the factors affecting the incremental maintenance, and finally we study the maintenance limit for each operator and for a combination of the operators.

Incremental view maintenance techniques are required for many new types of data models that are being increasingly used in industry. One of these models is the nested relational model that is used in the modelling of complex objects in databases. In this paper, we derive a group of expressions for incrementally evaluating query expressions in the nested relational model. We also present an algorithm to propagate base relation updates to a materialized view when the view is defined as a complex query.