Treatment effect estimation helps answer questions, such as whether a specific treatment affects the outcome of interest. One fundamental issue in this research is to alleviate the treatment assignment bias among those treated units and controlled units. Classical causal inference methods resort to the propensity score estimation, which unfortunately tends to be misspecified when only limited overlapping exists between the treated and the controlled units. Moreover, existing supervised methods mainly consider the treatment assignment information underlying the factual space, and thus, their performance of counterfactual inference may be degraded due to overfitting of the factual results. To alleviate those issues, we build on the optimal transport theory and propose a novel causal optimal transport (CausalOT) model to estimate an individual treatment effect (ITE). With the proposed propensity measure, CausalOT can infer the counterfactual outcome by solving a novel regularized optimal transport problem, which allows the utilization of global information on observational covariates to alleviate the issue of limited overlapping. In addition, a novel counterfactual loss is designed for CausalOT to align the factual outcome distribution with the counterfactual outcome distribution. Most importantly, we prove the theoretical generalization bound for the counterfactual error of CausalOT. Empirical studies on benchmark datasets confirm that the proposed CausalOT outperforms state-of-the-art causal inference methods.

Many regression tasks encounter an asymmetric distribution of information between training and testing phases where the additional information available in training, the so-called privileged information (PI), is often inaccessible in testing. In practice, the privileged information in training data might be expressed in different formats, such as continuous, ordinal, or binary values. However, most the existing learning using privileged information (LUPI) paradigms primarily deal with the continuous form of PI, preventing them from managing variational PI, which motivates this research. Therefore, in this paper, we propose a unified framework to systematically address the aforementioned three forms of privileged information. The proposed V-SVR+ method integrates continuous, ordinal, and binary PI into the learning process of support vector regression (SVR) via three losses. For continuous privileged information, we define a linear correcting (slack) function in the privileged information space to estimate slack variables in the standard SVR method using privileged information. For the ordinal relations of privileged information, we first rank the privileged information and then, regard this ordinal privileged information as auxiliary information used in the learning process of the SVR model. For the binary or Boolean privileged information, we infer a probabilistic dependency between the privileged information and labels from the summarized privileged information knowledge. Then, we transfer the privileged information knowledge to constraints and form a constrained optimization problem. We evaluate the proposed method in three applications: music emotion recognition from songs with the help of implicit information about music elements judged by composers; multiple object recognition from images with the help of implicit information about the object’s importance conveyed by the list of manually annotated image tags; and photo aesthetic assessment enhanced by high-level aesthetic attributes hidden in photos. Experiment results demonstrate that the proposed methods are superior to the classic learning paradigm when solving practical problems.

Abstract Causal inference is capable of estimating the treatment effect (i.e., the causal effect oftreatmenton theoutcome) to benefit the decision making in various domains. One fundamental challenge in this research is that the treatment assignment bias in observational data. To increase the validity of observational studies on causal inference, representation-based methods as the state-of-the-art have demonstrated the superior performance of treatment effect estimation. Most representation-based methods assume all observed covariates are pre-treatment (i.e., not affected by the treatment) and learn a balanced representation from these observed covariates for estimating treatment effect. Unfortunately, this assumption is often too strict a requirement in practice, as some covariates are changed by doing an intervention on treatment (i.e., post-treatment). By contrast, the balanced representation learned from unchanged covariates thus biases the treatment effect estimation. In light of this, we propose a deep treatment-adaptive architecture (DTANet) that can address the post-treatment covariates and provide a unbiased treatment effect estimation. Generally speaking, the contributions of this work are threefold. First, our theoretical results guarantee DTANet can identify treatment effect from observations. Second, we introduce a novel regularization of orthogonality projection to ensure that the learned confounding representation is invariant and not being contaminated by the treatment, meanwhile mediate variable representation is informative and discriminative for predicting the outcome. Finally, we build on the optimal transport and learn a treatment-invariant representation for the unobserved confounders to alleviate the confounding bias.

Underwriting is an important process in insurance and is concerned with accepting individuals into insurance policy with tolerable claim risk. Underwriting is a tedious and labor intensive process relying on underwriters' domain knowledge and experience, thus is labor intensive and prone to error. Machine learning models are recently applied to automate the underwriting process and thus to ease the burden on the underwriters as well as improve underwriting accuracy. However, observational data used for underwriting modelling is high dimensional, sparse and incomplete, due to the dynamic evolving nature (e.g., upgrade) of business information systems. Simply applying traditional supervised learning methods e.g., logistic regression or Gradient boosting on such highly incomplete data usually leads to the unsatisfactory underwriting result, thus requiring practical data imputation for training quality improvement. In this paper, rather than choosing off-the-shelf solutions tackling the complex data missing problem, we propose an innovative Generative Adversarial Nets (GAN) framework that can capture the missing pattern from a causal perspective. Specifically, we design a structural causal model to learn the causal relations underlying the missing pattern of data. Then, we devise a Causality-aware Generative network (CaGen) using the learned causal relationship prior to generating missing values, and correct the imputed values via the adversarial learning. We also show that CaGen significantly improves the underwriting prediction in real-world insurance applications.

Harnessing data from social media to monitor health events is a promising avenue for public health surveillance. A key step is the detection of reports of a disease (referred to as ‘health mention classification’) amongst tweets that mention disease words. Prior work shows that figurative usage of disease words may prove to be challenging for health mention classification. Since the experience of a disease is associated with a negative sentiment, we present a method that utilises sentiment information to improve health mention classification. Specifically, our classifier for health mention classification combines pre-trained contextual word representations with sentiment distributions of words in the tweet. For our experiments, we extend a benchmark dataset of tweets for health mention classification, adding over 14k manually annotated tweets across diseases. We also additionally annotate each tweet with a label that indicates if the disease words are used in a figurative sense. Our classifier outperforms current SOTA approaches in detecting both health-related and figurative tweets that mention disease words. We also show that tweets containing disease words are mentioned figuratively more often than in a health-related context, proving to be challenging for classifiers targeting health-related tweets.

Multi-relation embedding is a popular approach to knowledge base completion that learns embedding representations of entities and relations to compute the plausibility of missing triplet. The effectiveness of embedding approach depends on the sparsity of KB and falls for infrequent entities that only appeared a few times. This paper addresses this issue by proposing a new model exploiting the entity-independent transitive relation patterns, namely Transitive Relation Embedding (TRE). The TRE model alleviates the sparsity problem for predicting on infrequent entities while enjoys the generalisation power of embedding. Experiments on three public datasets against seven baselines showed the merits of TRE in terms of knowledge base completion accuracy as well as computational complexity.

Summary Jaccard Similarity has been widely used to measure the distance between two sets (or preference profiles) owned by two different users. Yet, in the private data collection scenario, it requires the untrusted curator could only estimate an approximately accurate Jaccard similarity of the involved users but without being allowed to access their preference profiles. This paper aims to address the above requirements by considering the local differential privacy model. To achieve this, we initially focused on a particular hash technique, MinHash, which was originally invented to estimate the Jaccard similarity efficiently. We designed the PrivMin algorithm to achieve the perturbation of MinHash signature by adopting Exponential mechanism and build the Locally Differentially Private Jaccard Similarity Estimation (LDP‐JSE) protocol for allowing the untrusted curator to approximately estimate Jaccard similarity. Theoretical and empirical results demonstrate that the proposed protocol can retain a highly acceptable utility of the estimated similarity as well as preserving privacy.

We build an ubiquitous abnormal activity detection system, namely NotiFi, for accurately detecting the abnormal activities on commercial off-the-shelf (COTS) IEEE 802.11 devices. In contrast to the traditional wearable sensor based and computer vision based systems which require additional sensors or enough lighting in line-of-sight (LoS) scenario, we proceed directly with abnormal activity characterization and activity modeling at the WiFi signal level based on Channel State Information (CSI). The intuition of NotiFi is that whenever the human body occludes the wireless signal transmitting from the access point to the receiver, the phase and the amplitude information of Channel State Information (CSI) will change sensitively. By creating a multiple hierarchical Dirichlet processes, NotiFi automatically learns the number of human body activity categories for abnormal detection. Experimental results in three typical indoor environments indicate that NotiFi can achieve satisfactory performance in accuracy, robustness and stability.

Sleep stage identification is the first step in modern sleep disorder diagnostics process. K-complex is an indicator for the sleep stage 2. However, due to the ambiguity of the translation of the medical standards into a computer-based procedure, reliability of automated K-complex detection from the EEG wave is still far from expectation. More specifically, there are some significant barriers to the research of automatic K-complex detection. First, there is no adequate description of K-complex that makes it difficult to develop automatic detection algorithm. Second, human experts only provided the label for whether a whole EEG segment contains K-complex or not, rather than individual labels for each subsegment. These barriers render most pattern recognition algorithms inapplicable in detecting K-complex. In this paper, we attempt to address these two challenges, by designing a new feature extraction method that can transform visual features of the EEG wave with any length into mathematical representation and proposing a hybrid-synergic machine learning method to build a K-complex classifier. The tenfold cross-validation results indicate that both the accuracy and the precision of this proposed model are at least as good as a human expert in K-complex detection.