Abstract--Deep Neural Networks (DNNs) have been extensively used in many areas including image processing, medical diagnostics, and autonomous driving. However, DNNs can exhibit erroneous behaviours that may lead to critical errors, especially when used in safety-critical systems. Inspired by testing techniques for traditional software systems, researchers have proposed neuron coverage criteria, as an analogy to source code coverage, to guide the testing of DNN models. Despite very active research on DNN coverage, several recent studies have questioned the usefulness of such criteria in guiding DNN testing. Further, from a practical standpoint, these criteria are white-box as they require access to the internals or training data of DNN models, which is in many contexts not feasible or convenient. In this paper, we investigate black-box input diversity metrics as an alternative to white-box coverage criteria. To this end, we first select and adapt three diversity metrics and study, in a controlled manner, their capacity to measure actual diversity in input sets. We further compare diversity with state-of-the-art white-box coverage criteria. Our experiments show that relying on the diversity of image features embedded in test input sets is a more reliable indicator than coverage criteria to effectively guide the testing of DNNs. Indeed, we found that one of our selected black-box diversity metrics far outperforms existing coverage criteria in terms of fault-revealing capability and computational time. Results also confirm the suspicions that state-of-the-art coverage metrics are not adequate to guide the construction of test input sets to detect as many faults as possible with natural inputs. Over the last decade, Deep Neural Networks (DNNs) In fact, in traditional software systems, testers rely on have achieved great performance in many domains, such coverage metrics as they assume that (1) inputs covering as image processing [1], [2], medical diagnostics [3], [4], [5], the same part of the source code are homogeneous, i.e, speech recognition [6] and autonomous driving [7], [8]. However these assumptions break down in critical errors. Therefore, like traditional software, DNN testing as (1) as opposed to code coverage, neuron DNNs need to be tested effectively to ensure their reliability coverage does not necessarily fully exercise the implicit logic and safety. While full coverage does not ensure to validate their proposed criteria [12], [13], [14], [10], [11].

This paper considers the problem of learning history dependent task specifications, e.g. automata and temporal logic, from expert demonstrations. Unfortunately, the (countably infinite) number of tasks under consideration combined with an a-priori ignorance of what historical features are needed to encode the demonstrated task makes existing approaches to learning tasks from demonstrations inapplicable. To address this deficit, we propose Demonstration Informed Specification Search (DISS): a family of algorithms parameterized by black box access to (i) a maximum entropy planner and (ii) an algorithm for identifying concepts, e.g., automata, from labeled examples. DISS works by alternating between (i) conjecturing labeled examples to make the demonstrations less surprising and (ii) sampling concepts consistent with the current labeled examples. In the context of tasks described by deterministic finite automata, we provide a concrete implementation of DISS that efficiently combines partial knowledge of the task and a single expert demonstration to identify the full task specification.

With the development of AIoT, data-driven attack detection methods for cyber-physical systems (CPSs) have attracted lots of attention. However, existing methods usually adopt tractable distributions to approximate data distributions, which are not suitable for complex systems. Besides, the correlation of the data in different channels does not attract sufficient attention. To address these issues, we use energy-based generative models, which are less restrictive on functional forms of the data distribution. In addition, graph neural networks are used to explicitly model the correlation of the data in different channels. In the end, we propose TFDPM, a general framework for attack detection tasks in CPSs. It simultaneously extracts temporal pattern and feature pattern given the historical data. Then extract features are sent to a conditional diffusion probabilistic model. Predicted values can be obtained with the conditional generative network and attacks are detected based on the difference between predicted values and observed values. In addition, to realize real-time detection, a conditional noise scheduling network is proposed to accelerate the prediction process. Experimental results show that TFDPM outperforms existing state-of-the-art attack detection methods. The noise scheduling network increases the detection speed by three times.

Recent advancements in self-supervised learning have reduced the gap between supervised and unsupervised representation learning. However, most self-supervised and deep clustering techniques rely heavily on data augmentation, rendering them ineffective for many learning tasks where insufficient domain knowledge exists for performing augmentation. We propose a new self-distillation based algorithm for domain-agnostic clustering. Our method builds upon the existing deep clustering frameworks and requires no separate student model. The proposed method outperforms existing domain agnostic (augmentation-free) algorithms on CIFAR-10. We empirically demonstrate that knowledge distillation can improve unsupervised representation learning by extracting richer `dark knowledge' from the model than using predicted labels alone. Preliminary experiments also suggest that self-distillation improves the convergence of DeepCluster-v2.

We'll divide this story into four sections: These are those data samples that look like normal samples but are perturbed in a certain way to fool the machine learning systems. For instance, in a given image not all the pixels are of the same importance, if you could identify the most important pixels (for the ML system to make a classification decision) and change them, your algorithm decision will also change and your sample will still appear to be normal. Why does the adversarial attack happen? Before we point out the reason as to why they happen, we just want to let you know that no ML algorithm is safe from adversarial attacks be it Logistic regression, softmax regression, SVM, Decision tree, Nearest neighbors, or deep learning models. Adversarial examples happen because of the excessive linearity in the systems.

A key assumption in supervised learning is that training and test data follow the same probability distribution. However, this fundamental assumption is not always satisfied in practice, e.g., due to changing environments, sample selection bias, privacy concerns, or high labeling costs. Transfer learning (TL) relaxes this assumption and allows us to learn under distribution shift. Classical TL methods typically rely on importance-weighting -- a predictor is trained based on the training losses weighted according to the importance (i.e., the test-over-training density ratio). However, as real-world machine learning tasks are becoming increasingly complex, high-dimensional, and dynamical, novel approaches are explored to cope with such challenges recently. In this article, after introducing the foundation of TL based on importance-weighting, we review recent advances based on joint and dynamic importance-predictor estimation. Furthermore, we introduce a method of causal mechanism transfer that incorporates causal structure in TL. Finally, we discuss future perspectives of TL research.

The mental disorder of online users is determined using social media posts. The major challenge in this domain is to avail the ethical clearance for using the user generated text on social media platforms. Academic re searchers identified the problem of insufficient and unlabeled data for mental health classification. To handle this issue, we have studied the effect of data augmentation techniques on domain specific user generated text for mental health classification. Among the existing well established data augmentation techniques, we have identified Easy Data Augmentation (EDA), conditional BERT, and Back Translation (BT) as the potential techniques for generating additional text to improve the performance of classifiers. Further, three different classifiers Random Forest (RF), Support Vector Machine (SVM) and Logistic Regression (LR) are employed for analyzing the impact of data augmentation on two publicly available social media datasets. The experiments mental results show significant improvements in classifiers performance when trained on the augmented data.

Research on knowledge graph (KG) construction, completion, inference, and applications has grown rapidly in recent years since it offers a powerful tool for modeling human knowledge in graph forms. Nodes in KGs denote entities and links represent relations between entities. The basic building blocks of KG are entity-relation triples in form of (subject, predicate, object) introduced by the Resource Description Framework (RDF). Learning representations for entities and relations in low dimensional vector spaces is one of the most active research topics in the field. Entity type offers a valuable piece of information to KG learning tasks. Better results in KG-related tasks have been achieved with the help of entity type. For example, TKRL [1] uses a hierarchical type encoder for KG completion by incorporating entity type information. AutoETER [2] adopts a similar approach but encodes the type information with projection matrices. Based on DistMult [3] and ComplEx [4] embedding, [5] propose an improved factorization model without explicit type supervision.

The development of high-throughput sequencing technologies has led to the production of largescale molecular profiling data, allowing us to gain insights into underlying biological processes (Wid lak, 2013). One such technology is microarray sequencing, in which mRNA counts are used to describe gene expression. Such data, known as transcriptomics, are widely used in the biomedical domain and when analyzed alongside survival times have provided extraordinary opportunities for biomarker characterization and prognostic modelling (Bøvelstad et al., 2007; Lloyd et al., 2015; Lightbody et al., 2019; Lu et al., 2021). However, profiling data is often high-dimensional, which introduces several statistical challenges including: (i) variable selection, (ii) effect estimation of the features, and (iii) scalable computation. The task of variable selection is particularly important, as few genes typically have an effect on the outcome. Motivated by clinical applicability, we propose a state-of-the-art scalable (variational) Bayesian variable selection method for the proportional hazards models. In recent years, several methods have been proposed to analyze sparse high-dimensional data, with one of the most popular being the LASSO (Tibshirani, 1996). As biomedical studies are often concerned with clinical phenotypes, such as time to disease recurrence or overall survival time, these methods have been adapted to support survival analysis (Antoniadis et al., 2010; Witten and Tibshirani, 2010). For instance, the LASSO, ridge and elastic-net penalties have all been extended to the proportional hazards model (Tibshirani, 1997; Gui and Li, 2005; Zou and Hastie, 2005; Simon et al., 2011).

When one observes a sequence of variables $(x_1, y_1), ..., (x_n, y_n)$, conformal prediction is a methodology that allows to estimate a confidence set for $y_{n+1}$ given $x_{n+1}$ by merely assuming that the distribution of the data is exchangeable. While appealing, the computation of such set turns out to be infeasible in general, e.g. when the unknown variable $y_{n+1}$ is continuous. In this paper, we combine conformal prediction techniques with algorithmic stability bounds to derive a prediction set computable with a single model fit. We perform some numerical experiments that illustrate the tightness of our estimation when the sample size is sufficiently large.