Abstract Due to the scarcity of anomalies, deep anomaly detection models are typically trained in an unsupervised or semi-supervised manner, depending on the availability of a small number of labeled samples. Currently, most unsupervised approaches detect anomalies by identifying the deviant patterns from normal samples, and some semi-supervised studies also use labeled anomalies to improve performance. However, few studies have focused on how to take advantage of potential anomalies in an easily obtained and large-scale unlabeled dataset. Meanwhile, in a semi-supervised setting, although we assume there will be a small number of labeled anomalies, the task of anomaly classification is under-exploited, which is important for domain experts. In this work, we focus on the problem of anomaly detection and classification with limited labeled samples and a large number of unlabeled samples. To this end, we develop a few-shot anomaly detection and classification model based on reinforced data selection with a combinatorial reward, called FADScr. FADScr iteratively improves performance by exploring the unlabeled dataset and selects informative samples to augment the training set to enhance both anomaly detection and classification. Experimental results show that our proposed framework is able to improve the performance of anomaly detection and classification with only a few labeled samples initially.

Capturing skeleton data is an important area of computer vision, especially for use in a virtual reality (VR) setting. As new techniques to extract skeletons come out, the popularity of skeleton-based motion data has increased. While the skeleton data appears to be anonymous, it can be exploited to discover personally identifiable information (PII). This poses a risk of unintentional privacy leakages when skeletons are publicly displayed, like in a VR environment. We explore the privacy implications posed by the skeleton data, focusing on the privacy and utility trade-off and current privacy-preserving techniques. In this paper, we propose a new baseline attack model Linkage Attack Neural Network (LAN) that acts as a matching classifier to distinguish whether two skeleton-sequences are from the same actor. Then we propose a new defense model, the Privacy-centric Deep Motion Retargeting (PMR) model that is an adversarial/cooperatively trained motion retargeting model. Additionally, we incorporate explanation techniques to identify and mask the most privacy-sensitive joints, either by zeroing them out or adding controlled noise. Finally, we propose a transformer-based motion retargeting model designed for real-time applications, leveraging autoregressive decoding for continuous, frame-by-frame skeleton generation.

Pre-trained language models (PLMs) are commonly used for various downstream natural language processing tasks via fine-tuning. However, recent studies have demonstrated that PLMs are vulnerable to backdoor attacks, which can mislabel poisoned samples to target outputs even after a vanilla fine-tuning process. The key challenge for defending against the backdoored PLMs is that end users who adopt the PLMs for their downstream tasks usually do not have any knowledge about the attacking strategies, such as triggers. To tackle this challenge, in this work, we propose a backdoor mitigation approach, PURE, via head pruning and normalization of attention weights. The idea is to prune the attention heads that are potentially affected by poisoned texts with only clean texts on hand and then further normalize the weights of remaining attention heads to mitigate the backdoor impacts. We conduct experiments to defend against various backdoor attacks on the classification task. The experimental results show the effectiveness of PURE in lowering the attack success rate without sacrificing the performance on clean texts.

The recent California Consumer Privacy Act (CCPA) requires that personal data shall be limited to what is necessary for business purposes. Business services shall “implement technical safeguards that prohibit re-identification of the consumer to whom the information may pertain”. For recommender systems, we believe the legal concepts of limitation and technical safeguard are not specific enough to operationalize in practice. This study makes efforts to map the legislative challenges to practice of reducing personal data. More importantly, we borrowed the notion of uncertainty from the machine learning community, and added it as another aspect of recommendation utility, in addition to recommendation accuracy, to guide the data reduction process. The benefit of using uncertainty is that we have more comprehensive consideration while reducing the personal data. In addition, two major types of uncertainty in machine learning models: aleatoric uncertainty and epistemic uncertainty, helped us formulate two groups of data reduction strategies: within-user and between-user. We conducted a series of analyses regarding uncertainty change and accuracy loss caused by different data reduction strategies. We found that at the aggregate level, data reduction is feasible with certain data reduction strategies. At the individual level, the recommendation utility (both uncertainty and accuracy) loss incurred by data reduction disparately impacts different users — a finding which has implications for fairness and transparency of AI models. Our results reveal the difficulty and intricacy of the data reduction problem in the context of recommender systems.

Skeleton-based motion capture and visualization is an important computer vision task, especially in the virtual reality (VR) environment. It has grown increasingly popular due to the ease of gathering skeleton data and the high demand of virtual socialization. The captured skeleton data seems anonymous but can still be used to extract personal identifiable information (PII). This can lead to an unintended privacy leakage inside a VR meta-verse. We propose a novel linkage attack on skeleton-based motion visualization. It detects if a target and a reference skeleton are the same individual. The proposed model, called Linkage Attack Neural Network (LAN), is based on the principles of a Siamese Network. It incorporates deep neural networks to embed the relevant PII then uses a classifier to match the reference and target skeletons. We also employ classical and deep motion retargeting (MR) to cast the target skeleton onto a dummy skeleton such that the motion sequence is anonymized for privacy protection. Our evaluation shows that the effectiveness of LAN in the linkage attack and the effectiveness of MR in anonymization.

Sequential anomaly detection has received more and more attention because of its wide applications in various domains, such as debugging system failures via logs. Researchers have recently proposed many deep learning-based approaches for sequential anomaly detection. However, these approaches work as black-boxed models, not providing explanations for detected anomalies. On the other hand, explainability is a critical requirement to build trustworthiness in detection results. Moreover, domain experts would like to learn why a sequence is labeled as an anomaly. To overcome this challenge, in this paper, we propose a framework for Explainable Sequential Anomaly Detection (ESAD) in a semi-supervised setting. As there are various normal and abnormal behaviors in sequential data, ESAD derives multiple prototypes to describe diverse normal and abnormal sequences. Each prototype can encode one type of normal or abnormal behavior. Given a new sequence, if the sequence is similar to an abnormal prototype, the sequence will be detected as abnormal. After decoding the abnormal prototype as a prototypical sequence, domain experts can further understand the newly detected abnormal sequence by examining the prototypical sequence. We conduct experiments on one log dataset and two text datasets. Experimental results including quantitative and qualitative analysis on three datasets show the effectiveness of our model.

Sequential anomaly detection has been studied for decades because of its wide spectrum of applications and obtained significant improvement in recent years by utilizing deep learning techniques. As an increasing number of anomaly detection models are applied to high-stake tasks involving human beings, it is critical to understand the reasons why the samples are labeled as anomalies. In this work, we propose a Globally and Locally Explainable Anomaly Detection (GLEAD) framework targeting sequential data. Especially, considering that the anomalies are usually diverse, we make use of the multi-head self-attention techniques to derive representations for sequences as well as prototypes, which capture a variety of patterns in anomalies. The attention mechanism highlights the abnormal entries with high attention weights in the abnormal sequences for the local explanation. Moreover, the prototypes of anomalies encoding the common patterns of abnormal sequences are derived to achieve the global explanation. Experimental results on two sequential anomaly detection datasets show that our approach can detect abnormal sequences and provide local and global explanations.

Due to the scarcity of anomalies, deep anomaly detection models are predominately trained in an unsupervised or semi-supervised manner depending on the availability of a small number of labeled samples. Currently, most unsupervised approaches detect anomalies by identifying the deviate patterns, and some semi-supervised studies also use labeled anomalies to improve performance. However, few studies have focused on how to take advantage of potential anomalies in an easily obtained and large-scale unlabeled dataset. Meanwhile, in a semi-supervised setting, although we assume having a small number of labeled anomalies, the task of anomaly classification is under-exploited. In this work, considering the problem of anomaly detection and classification by giving limited labeled samples as well as a large number of unlabeled samples, we propose a few-shot anomaly detection and classification model through reinforced data selection (FADS), a novel framework that iteratively improves the performance of anomaly detection and classification by exploring the unlabeled dataset to augment the training set. Experimental results show that FADS is able to improve the performance of anomaly detection and classification with only a few labeled samples initially.

Anomaly detection in sequential log data is a common data analysis task as it contributes to detecting critical information, such as malfunctions of systems. However, due to the scarcity of anomalies, the traditional supervised learning approaches cannot be applied for anomaly detection tasks. Meanwhile, most of the existing studies only focus on identifying the anomalous log sequences and cannot further detect the anomalous events in a sequence. In this work, we present InterpretableSAD, an interpretable log anomaly detection framework that can achieve both anomalous sequence and fine-grained event detection. Given a set of normal log sequences, we propose a data augmentation strategy to generate a set of anomalous sequences via negative sampling so that we can train a binary classification model based on the observed normal sequences and the generated anomalous sequences. After training, the classification model is able to detect real anomalous log sequences. We then consider the anomalous event detection as a model interpretation problem and apply an interpretable machine learning technique in a novel way to detect which parts of the sequences, a.k.a, anomalous events, lead to anomalous issues. Experimental results on three log datasets show the effectiveness of our proposed framework.

Preserving differential privacy has been well studied under the centralized setting. However, it’s very challenging to preserve differential privacy under multiparty setting, especially for the vertically partitioned case. In this work, we propose a new framework for differential privacy preserving multiparty learning in the vertically partitioned setting. Our core idea is based on the functional mechanism that achieves differential privacy of the released model by adding noise to the objective function. We show the server can simply dissect the objective function into single-party and cross-party sub-functionsa, and allocate computation and perturbation of their polynomial coefficients to local parties. Our method needs only one round of noise addition and secure aggregation. The released model in our framework achieves the same utility as applying the functional mechanism in the centralized setting. Evaluation on real-world and synthetic datasets for linear and logistic regressions shows the effectiveness of our proposed method.

In differentially private stochastic gradient descent (DPSGD), gradient clipping and random noise addition disproportionately affect underrepresented and complex classes and subgroups. As a consequence, DPSGD has disparate impact: the accuracy of a model trained using DPSGD tends to decrease more on these classes and subgroups vs. the original, non-private model. If the original model is unfair in the sense that its accuracy is not the same across all subgroups, DPSGD exacerbates this unfairness. In this work, we study the inequality in utility loss due to differential privacy, which compares the changes in prediction accuracy w.r.t. each group between the private model and the non-private model. We analyze the cost of privacy w.r.t. each group and explain how the group sample size along with other factors is related to the privacy impact on group accuracy. Furthermore, we propose a modified DPSGD algorithm, called DPSGD-F, to achieve differential privacy, equal costs of differential privacy, and good utility. DPSGD-F adaptively adjusts the contribution of samples in a group depending on the group clipping bias such that differential privacy has no disparate impact on group accuracy. Our experimental evaluation shows the effectiveness of our removal algorithm on achieving equal costs of differential privacy with satisfactory utility.

Federated learning is an emerging framework that builds centralized machine learning models with training data distributed across multiple devices.Most of the previous works about federated learning focus on the privacy protection and communication cost reduction.However, how to achieve fairness in federated learning is underexplored and challenging especially when testing data distribution is different from training distribution or even unknown.Introducing simple fairness constraints on the centralized model cannot achieve model fairness on unknown testing data.In this paper, we develop a fairness-aware agnostic federated learning framework (Agnostic-Fair) to deal with the challenge of unknown testing distribution.We use kernel reweighing functions to assign a reweighing value on each training sample in both loss function and fairness constraint.Therefore, the centralized model built from AgnosticFair can achieve high accuracy and fairness guarantee on unknown testing data.Moreover, the built model can be directly applied to local sites as it guarantees fairness on local data distributions.To our best knowledge, this is the first work to achieve fairness in federated learning.Experimental results on two real datasets demonstrate the effectiveness in terms of both utility and fairness under data shift scenarios.

How to achieve fairness is important for next generation machine learning. Two tasks that are equally important in fair machine learning are how to obtain fair datasets and how to build fair classifiers. In this work, we propose a new generative adversarial network (GAN) model for fair machine learning, named FairGAN+. FairGAN+contains a generator to generate close-to-real samples, a classifier to predict class labels and three discriminators to assist adversarial learning. FairGAN+simultaneously achieves fair data generation and classification by co-training the generative model and the classifier through joint adversarial games with the discriminators. Evaluations on real world data show the effectiveness of FairGAN+on both fair data generation and fair classification.

Achieving fairness in learning models is currently an imperative task in machine learning. Meanwhile, recent research showed that fairness should be studied from the causal perspective, and proposed a number of fairness criteria based on Pearl's causal modeling framework. In this paper, we investigate the problem of building causal fairness-aware generative adversarial networks (CFGAN), which can learn a close distribution from a given dataset, while also ensuring various causal fairness criteria based on a given causal graph. CFGAN adopts two generators, whose structures are purposefully designed to reflect the structures of causal graph and interventional graph. Therefore, the two generators can respectively simulate the underlying causal model that generates the real data, as well as the causal model after the intervention. On the other hand, two discriminators are used for producing a close-to-real distribution, as well as for achieving various fairness criteria based on causal quantities simulated by generators. Experiments on a real-world dataset show that CFGAN can generate high quality fair data.

Fairness-aware learning is increasingly important in data mining. Discrimination prevention aims to prevent discrimination in the training data before it is used to conduct predictive analysis. In this paper, we focus on fair data generation that ensures the generated data is discrimination free. Inspired by generative adversarial networks (GAN), we present fairness-aware generative adversarial networks, called FairGAN, which are able to learn a generator producing fair data and also preserving good data utility. Compared with the naive fair data generation models, FairGAN further ensures the classifiers which are trained on generated data can achieve fair classification on real data. Experiments on a real dataset show the effectiveness of FairGAN.