Multi-view learning methods often focus on improving decision accuracy while neglecting the decision uncertainty, which significantly restricts their applications in safety-critical scenarios. To address this, trusted multi-view learning methods estimate prediction uncertainties by learning class distributions from each instance. However, these methods heavily rely on high-quality ground-truth labels. This motivates us to delve into a new problem: how to develop a reliable multi-view learning model under the guidance of noisy labels? We propose the Trusted Multi-view Noise Refining (TMNR) method to address this challenge by modeling label noise arising from low-quality data features and easily-confused classes. TMNR employs evidential deep neural networks to construct view-specific opinions that capture both beliefs and uncertainty. These opinions are then transformed through noise correlation matrices to align with the noisy supervision, where matrix elements are constrained by sample uncertainty to reflect label reliability. Furthermore, considering the challenge of jointly optimizing the evidence network and noise correlation matrices under noisy supervision, we further propose Trusted Multi-view Noise Re-Refining (TMNR$^{\mathbf{2}}$2), which disentangles this complex co-training problem by establishing different training objectives for distinct modules. TMNR$^{\mathbf{2}}$2 identifies potentially mislabeled samples through evidence-label consistency and generates pseudo-labels from neighboring information. By assigning clean samples to optimize evidential networks and noisy samples to guide noise correlation matrices, respectively, TMNR$^{\mathbf{2}}$2 reduces mapping interference and achieves stabilized training. We empirically evaluate our methods against state-of-the-art baselines on 7 multi-view datasets. Experimental results demonstrate that TMNR$^{\mathbf{2}}$2 significantly outperforms baseline methods, with average accuracy improvements of 7% on datasets with 50% label noise.

Autonomous and semi-autonomous systems are using deep learning models to improve decision-making. However, deep classifiers can be overly confident in their incorrect predictions, a major issue especially in safety-critical domains. The present study introduces three foundational desiderata for developing real-world risk-aware classification systems. Expanding upon the previously proposed Evidential Deep Learning ( EDL ), we demonstrate the unity between these principles and EDL ’s operational attributes. We then augment EDL empowering autonomous agents to exercise discretion during structured decision-making when uncertainty and risks are inherent. We rigorously examine empirical scenarios to substantiate these theoretical innovations. In contrast to existing risk-aware classifiers, our proposed methodologies consistently exhibit superior performance, underscoring their transformative potential in risk-conscious classification strategies. • Evidential deep learning uses Dirichlet distributions to represent the predictive uncertainty of neural classifiers. • Pignistic probabilities can be used to model rational decision-making under uncertainty. • Risk awareness can be integrated into evidential classifiers using pignistic Dirichlet priors.

This full-day tutorial introduces modern techniques for practical uncertainty quantification specifically in the context of multi-class and multi-label text classification. First, we explain the usefulness of estimating aleatoric uncertainty and epistemic uncertainty for text classification models. Then, we describe several state-of-the-art approaches to uncertainty quantification and analyze their scalability to big text data: Virtual Ensemble in GBDT, Bayesian Deep Learning (including Deep Ensemble, Monte-Carlo Dropout, Bayes by Backprop, and their generalization Epistemic Neural Networks), Evidential Deep Learning (including Prior Networks and Posterior Networks), as well as Distance Awareness (including Spectral-normalized Neural Gaussian Process and Deep Deterministic Uncertainty). Next, we talk about the latest advances in uncertainty quantification for pre-trained language models (including asking language models to express their uncertainty, interpreting uncertainties of text classifiers built on large-scale language models, uncertainty estimation in text generation, calibration of language models, and calibration for in-context learning). After that, we discuss typical application scenarios of uncertainty quantification in text classification (including in-domain calibration, cross-domain robustness, and novel class detection). Finally, we list popular performance metrics for the evaluation of uncertainty quantification effectiveness in text classification. Practical hands-on examples/exercises are provided to the attendees for them to experiment with different uncertainty quantification methods on a few real-world text classification datasets such as CLINC150.

As Deep Neural Networks (DNN) make their way into safety-critical decision processes, it becomes imperative to have robust and reliable uncertainty estimates for their predictions for both in-distribution and out-of-distribution (OOD) examples. This is particularly important in real-life high-risk settings such as healthcare, where OOD examples (e.g., patients with previously unseen or rare labels, i.e., diagnoses) are frequent, and an incorrect clinical decision might put human life in danger, in addition to having severe ethical and financial costs. While evidential uncertainty estimates for deep learning have been studied for multi-class problems, research in multi-label settings remains untapped. In this paper, we propose a DEep Evidential Doctor (DEED), which is a novel deterministic approach to estimate multi-label targets along with uncertainty. We achieve this by placing evidential priors over the original likelihood functions and directly estimating the parameters of the evidential distribution using a novel loss function. Additionally, we build a redundancy layer (particularly for high uncertainty and OOD examples) to minimize the risk associated with erroneous decisions based on dubious predictions. We achieve this by learning the mapping between the evidential space and a continuous semantic label embedding space via a recurrent decoder. Thereby inferring, even in the case of OOD examples, reasonably close predictions to avoid catastrophic consequences. We demonstrate the effectiveness of DEED on a digit classification task based on a modified multi-label MNIST dataset, and further evaluate it on a diagnosis prediction task from a real-life electronic health record dataset. We highlight that in terms of prediction scores, our approach is on par with the existing state-of-the-art having a clear advantage of generating reliable, memory and time-efficient uncertainty estimates with minimal changes to any multi-label DNN classifier.

Many software systems, such as online social networks, enable users to share information about themselves. Although the action of sharing is simple, it requires an elaborate thought process on privacy: what to share, with whom to share, and for what purposes. Thinking about these for each piece of content to be shared is tedious. Recent approaches to tackle this problem build personal assistants that can help users by learning what is private over time and recommending privacy labels such as private or public to individual content that a user considers sharing. However, privacy is inherently ambiguous and highly personal . Existing approaches to recommend privacy decisions do not address these aspects of privacy sufficiently. Ideally, a personal assistant should be able to adjust its recommendation based on a given user, considering that user’s privacy understanding. Moreover, the personal assistant should be able to assess when its recommendation would be uncertain and let the user make the decision on her own. Accordingly, this article proposes a personal assistant that uses evidential deep learning to classify content based on its privacy label. An important characteristic of the personal assistant is that it can model its uncertainty in its decisions explicitly, determine that it does not know the answer, and delegate from making a recommendation when its uncertainty is high. By factoring in the user’s own understanding of privacy, such as risk factors or own labels, the personal assistant can personalize its recommendations per user. We evaluate our proposed personal assistant using a well-known dataset. Our results show that our personal assistant can accurately identify uncertain cases, personalize them to its user’s needs, and thus helps users preserve their privacy well.

When collaborating with an artificial intelligence (AI) system, we need to assess when to trust its recommendations. Suppose we mistakenly trust it in regions where it is likely to err. In that case, catastrophic failures may occur, hence the need for Bayesian approaches for reasoning and learning to determine the confidence (or epistemic uncertainty) in the probabilities of the queried outcome. Pure Bayesian methods, however, suffer from high computational costs. To overcome them, we revert to efficient and effective approximations. In this paper, we focus on techniques that take the name of evidential reasoning and learning from the process of Bayesian update of given hypotheses based on additional evidence. This paper provides the reader with a gentle introduction to the area of investigation, the up-to-date research outcomes, and the open questions still left unanswered.

There is strong interest in leveraging reinforcement learning (RL) for information retrieval (IR) applications including search, recommendation, and advertising. Just in 2020, the term "reinforcement learning" was mentioned in more than 60 different papers published by ACM SIGIR. It has also been reported that Internet companies like Google and Alibaba have started to gain competitive advantages from their RL-based search and recommendation engines. This full-day tutorial gives IR researchers and practitioners who have no or little experience with RL the opportunity to learn about the fundamentals of modern RL in a practical hands-on setting. Furthermore, some representative applications of RL in IR systems will be introduced and discussed. By attending this tutorial, the participants will acquire a good knowledge of modern RL concepts and standard algorithms such as REINFORCE and DQN. This knowledge will help them better understand some of the latest IR publications involving RL, as well as prepare them to tackle their own practical IR problems using RL techniques and tools. Please refer to the tutorial website (https://rl-starterpack.github.io/) for more information.

In this paper, we propose risk-calibrated evidential deep classifiers to reduce the costs associated with classification errors. We use two main approaches. The first is to develop methods to quantify the uncertainty of a classifier's predictions and reduce the likelihood of acting on erroneous predictions. The second is a novel way to train the classifier such that erroneous classifications are biased towards less risky categories. We combine these two approaches in a principled way. While doing this, we extend evidential deep learning with pignistic probabilities, which are used to quantify uncertainty of classification predictions and model rational decision making under uncertainty.We evaluate the performance of our approach on several image classification tasks. We demonstrate that our approach allows to (i) incorporate misclassification cost while training deep classifiers, (ii) accurately quantify the uncertainty of classification predictions, and (iii) simultaneously learn how to make classification decisions to minimize expected cost of classification errors.

Deep neural networks are often ignorant about what they do not know and overconfident when they make uninformed predictions. Some recent approaches quantify classification uncertainty directly by training the model to output high uncertainty for the data samples close to class boundaries or from the outside of the training distribution. These approaches use an auxiliary data set during training to represent out-of-distribution samples. However, selection or creation of such an auxiliary data set is non-trivial, especially for high dimensional data such as images. In this work we develop a novel neural network model that is able to express both aleatoric and epistemic uncertainty to distinguish decision boundary and out-of-distribution regions of the feature space. To this end, variational autoencoders and generative adversarial networks are incorporated to automatically generate out-of-distribution exemplars for training. Through extensive analysis, we demonstrate that the proposed approach provides better estimates of uncertainty for in- and out-of-distribution samples, and adversarial examples on well-known data sets against state-of-the-art approaches including recent Bayesian approaches for neural networks and anomaly detection methods.

Efficient second-order probabilistic inference in uncertain Bayesian networks was recently introduced. However, such second -order inference methods presume training over complete training data. While the expectation-maximization framework is well-established for learning Bayesian network parameters for incomplete training data, the framework does not determine the covariance of the parameters. This paper introduces two methods to compute the covariances for the parameters of Bayesian networks or Markov random fields due to incomplete data for two-node networks. The first method computes the covariances directly from the posterior distribution of parameters, and the second method more efficiently estimates the covariances from the Fisher information matrix. Finally, the implications and effectiveness of these covariances is theoretically and empirically evaluated.

We enable aProbLog—a probabilistic logical programming approach—to reason in presence of uncertain probabilities represented as Beta-distributed random variables. We achieve the same performance of state-of-the-art algorithms for highly specified and engineered domains, while simultaneously we maintain the flexibility offered by aProbLog in handling complex relational domains. Our motivation is that faithfully capturing the distribution of probabilities is necessary to compute an expected utility for effective decision making under uncertainty: unfortunately, these probability distributions can be highly uncertain due to sparse data. To understand and accurately manipulate such probability distributions we need a well-defined theoretical framework that is provided by the Beta distribution, which specifies a distribution of probabilities representing all the possible values of a probability when the exact value is unknown.

editorial Free Access Share on Introduction to the Special Section on Trust and AI Editors: Jie Zhang NTU, Singapore NTU, SingaporeView Profile , Jamal Bentahar Concordia University, Canada Concordia University, CanadaView Profile , Rino Falcone ISTC-CNR, Italy ISTC-CNR, ItalyView Profile , Timothy J. Norman University of Southampton, UK University of Southampton, UKView Profile , Murat Şensoy Blue Prism Labs, UK Blue Prism Labs, UKView Profile Authors Info & Claims ACM Transactions on Internet TechnologyVolume 19Issue 4November 2019 Article No.: 44epp 1–3https://doi.org/10.1145/3365675Online:09 November 2019Publication History 0citation344DownloadsMetricsTotal Citations0Total Downloads344Last 12 Months114Last 6 weeks6 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 Alerts New Citation Alert!Please log in to your account Save to BinderSave to BinderCreate a New BinderNameCancelCreateExport CitationPublisher SiteView all FormatsPDF

In this paper we provide a critical analysis with metrics that will inform guidelines for designing distributed systems for Collective Situational Understanding (CSU). CSU requires both collective insight-i.e., accurate and deep understanding of a situation derived from uncertain and often sparse data and collective foresight-i.e., the ability to predict what will happen in the future. When it comes to complex scenarios, the need for a distributed CSU naturally emerges, as a single monolithic approach not only is unfeasible: it is also undesirable. We therefore propose a principled, critical analysis of AI techniques that can support specific tasks for CSU to derive guidelines for designing distributed systems for CSU.

This work analyzes the impact of correlated propositions when estimating the reporting behavior of information sources. These behavior estimates are critical for fusion, and traditional methods assume the propositions are statistically independent. A new source behavior estimation methods is presented that accounts for statistical dependencies between the training propositions. Simulations seem to indicate that the potential performance gains for accounting for the correlations is small relative to the increased computational complexity. One may conclude that the traditional independence assumption in source behavior estimation methods is reasonable even in cases where it is actually violated.

Distributed Denial of Service (DDoS) attacks continue to be one of the most severe threats in the Internet. The intrinsic challenge in preventing DDoS attacks is to distinguish them from legitimate flash crowds since two have many traffic characteristics in common. Today most DDoS detection techniques focus on finding parametric differences between the patterns in attack and legitimate traffic. However, such techniques are very sensitive to the threshold values set on the parameters and more importantly legitimate traffic features might be mimicked by smart attackers to generate requests that look like flash crowds. In this paper, we propose a framework for training networks for such smart attacks. Our framework is based on Deep Generative Network models and our contributions are two-fold.We first show that legitimate traffic features can be mimicked without explicitly modeling their distributions. Second, we introduce the concept of context-aware DDoS attacks. We show that an attacker can generate traffic that looks similar to flash crowds to be undetected for long periods of time. However, the ability of generating such attacks is constrained by the budget of the attacker. A context-aware attacker is the one that can intelligently use its budget to maximize the damage in the victim network. Our study provides a framework for training networks for such DDoS attack scenarios.

Deterministic neural nets have been shown to learn effective predictors on a wide range of machine learning problems. However, as the standard approach is to train the network to minimize a prediction loss, the resultant model remains ignorant to its prediction confidence. Orthogonally to Bayesian neural nets that indirectly infer prediction uncertainty through weight uncertainties, we propose explicit modeling of the same using the theory of subjective logic. By placing a Dirichlet distribution on the class probabilities, we treat predictions of a neural net as subjective opinions and learn the function that collects the evidence leading to these opinions by a deterministic neural net from data. The resultant predictor for a multi-class classification problem is another Dirichlet distribution whose parameters are set by the continuous output of a neural net. We provide a preliminary analysis on how the peculiarities of our new loss function drive improved uncertainty estimation. We observe that our method achieves unprecedented success on detection of out-of-distribution queries and endurance against adversarial perturbations.

Monitoring and following health and safety regulations are especially important - but made difficult - in hazardous work environments such as underground mines to prevent work place accidents and illnesses. Even though there are IoT solutions for health and safety, every work place has different characteristics and monitoring is typically done by humans in control rooms. During emergencies, conflicts may arise among prohibitions and obligations, and humans may not be better placed to make decision without any assistance as they do not have a bird's-eye-view of the environment. Motivated by this observations, in this paper, we discuss how health and safety regulations can be implemented using a semantic policy framework. We then show how this framework can be integrated into an in-use smart underground mine solution. We also evaluate the performance of our framework to show that it can cope with the complexity and the amount of data generated by the system.

With the proliferation of technology, connected and interconnected devices (henceforth referred to as IoT) are fast becoming a viable option to automate the day-to-day interactions of users with their environment—be it manufacturing or home-care automation. However, with the explosion of IoT deployments we have observed in recent years, manually governing the interactions between humans-to-devices—and especially devices-to- devices—is an impractical task, if not an impossible task. This is because devices have their own obligations and prohibitions in context, and humans are not equip to maintain a bird’s-eye-view of the interaction space. Motivated by this observation, in this paper, we propose an end-to-end framework that (a) automatically dis- covers devices, and their associated services and capabilities w.r.t. an ontology; (b) supports representation of high-level—and expressive—user policies to govern the devices and services in the environment; (c) pro- vides efficient procedur es to refine and reason about policies to automate the management of interactions; and (d) delegates similar capable devices to fulfill the interactions, when conflicts occur. We then present our initial work in instrumenting the framework and discuss its details.

Bootstrapping trust assessment where there is little or no evidence regarding a subject is a significant challenge for existing trust and reputation systems. When direct or indirect evidence is absent, existing approaches usually assume that all agents are equally trustworthy. This naive assumption makes existing approaches vulnerable to attacks such as Sybil and whitewashing. Inspired by real‐life scenarios, we argue that malicious agents may share some common patterns or complex features in their descriptions. If such patterns or features can be detected, they can be exploited to bootstrap trust assessments. Based on this idea, we propose the use of frequent subgraph mining and state‐of‐the‐art knowledge representation formalisms to estimate a priori trust for agents. Our approach first discovers significant patterns that may be used to characterize trustworthy and untrustworthy agents. Then, these patterns are used as features to train a regression model to estimate the trustworthiness of agents. Last, a priori trust for unknown agents (e.g., newcomers) is estimated using the discovered features based on the trained model. Through empirical evaluation, we show that the proposed approach significantly outperforms well‐known trust approaches if trustworthiness of agents is correlated with patterns in their descriptions or social networks. Furthermore, we show that the proposed approach performs at least as good as the existing approaches if such correlations do not exist.

As the semantic web expands, ontological data becomes distributed over a large network of data sources on the Web. Consequently, evaluating queries that aim to tap into this distributed semantic database necessitates the ability to consult multiple data sources efficiently. In this paper, we propose methods and heuristics to efficiently query distributed ontological data based on a series of properties of summarized data. In our approach, each source summarizes its data as another RDF graph, and relevant section of these summaries are merged and analyzed at query evaluation time. We show how the analysis of these summaries enables more efficient source selection, query pruning and transformation of expensive distributed joins into local joins.

Stereotypical trust modeling can be adopted by a buyer to effectively evaluate trustworthiness of a seller who has little or no past experience in e-marketplaces. The buyer forms trust stereotypes based on her past experience with other sellers. However, when the buyer has limited past experience with sellers, the formed stereotypes cannot accurately reflect her trust evaluation towards sellers. To address this issue, we propose a novel generalized stereotypical trust model. Specifically, we first build a semantic ontology to represent hierarchical relationships among seller attribute values. We then propose a fuzzy semantic decision tree (FSDT) learning method to construct trust stereotypes that generalizes over seller non-nominal attributes by splitting their values in a fuzzy manner, and generalizes over nominal attributes by replacing their specific values with more general terms according to the ontology. Experimental results confirm that our proposed model can more accurately measure the trustworthiness of sellers in simulated e-marketplaces where buyers have limited experience with sellers.

In coalition operations, information from different sources belong to different organisations have to be gathered and aggregated. The information from these resources may not be consistent. Inconsistencies in the gathered information creates severe uncertainties that hinders the usefulness of the information. In this paper, we have propose a Subjective Logic based approach for modelling the trustworthiness of information sources within a specific context. This model is used to handle inconsistencies through filtering information from less trustworthy sources.

In this paper, we propose a reputation mechanism for virtual marketplaces. The proposed approach is based on five-sense oriented feedback provision with the support of existing virtual reality technologies. We have conducted user studies to analyse users' attitude towards this new approach. These studies reveal that users prefer virtual marketplaces with our proposed reputation mechanism over that with traditional reputation mechanisms, and that our mechanism can effectively ensure user's trust in the virtual marketplaces and simultaneously promote user's trust in other users. Our approach is based on feedback from other users. Feedback from users could be very subjective and misleading for other users. Hence, we propose a novel mechanism to align subjective user feedback before reputation computations in virtual marketplaces. Results of our experiment demonstrate that with our feedback alignment approach, buyers can more accurately model sellers' reputation.

Many organizations depend on critical sensory information to achieve their tasks. As the number of those tasks increases, efficient determination and allocation of required resources in sensor networks become crucial. In this paper, we propose means to describe tasks semantically with their requirements and constraints so that software agents can reason about those tasks and determine what type of sensor resources they may need. Based on the semantic description and reasoning mechanisms, we propose a distributed agent-based approach to efficiently allocate sensor resources to tasks. Our evaluation of the proposed approach shows that not only it enables fully automated determination and allocation of resources for tasks, but also the resulting allocation is efficient and close to optimum.

The increasing number of service providers on the Web makes it challenging to select a provider for a specific service demand. Each service consumer has different expectations for a given service in different contexts, so the selection process should be consumer‐oriented and context‐dependent. Current approaches for service selection typically have consumers receive ratings of providers from other consumers, where the ratings reflect the consumers' overall subjective opinions. This may be misleading if consumers have different contexts and satisfaction criteria. In this paper, we propose that consumers objectively record their experiences, using an ontology to capture subtle details. This can then be interpreted by consumers according to their own criteria and contexts. We then integrate a method for addressing consumers who lie about their experiences, filtering them out during service selection. We demonstrate the value of our approach through experiments comparing our model with three recent rating‐based service selection models. Our experiments show that using the proposed approach, service consumers can select the service providers for their needs more accurately even if the consumers have different criteria, they change the contexts of their service demands over time, or a significant portion of them are liars.

This paper proposes an approach that enables agents to teach each other concepts from their ontologies using examples. Unlike other concept learning approaches, our approach enables the learner to elicit the most informative examples interactively from the teacher. Hence, the learner participates to the learning process actively. We empirically compare the proposed approach with the previous concept learning approaches. Our experiments show that using the proposed approach, agents can learn new concepts successfully and with fewer examples.

The unfair rating problem exists when a buying agent models the trustworthiness of selling agents by also relying on ratings of the sellers from other buyers. Different probabilistic approaches have been proposed to cope with this issue. In this paper, we first summarize these approaches and provide a detailed categorization of them. This includes our own "personalized" approach for addressing this problem. Based on the implication of such analysis, we then focus on experimental comparison of our approach with two key models in a framework that simulates a dynamic electronic marketplace environment. We specifically examine different scenarios, including ones where the majority of buyers are dishonest, buyers lack personal experience with sellers, sellers may vary their behavior, and buyers may provide a large number of ratings. Our study provides the basis for deciding which approach is most appropriate to employ, in which scenario.

Selecting the right parties to interact with is a fundamental problem in open and dynamic environments. The problem is amplified when the number of interacting parties is high, and the parties' reasons for selecting others vary. We examine the problem of service selection in an e-commerce setting where consumer agents cooperate to identify service providers that would satisfy their service needs the most. Previous approaches to service selection are usually based on capturing and exchanging the ratings of consumers to providers. Rating-based approaches have two major weaknesses. 1) ratings are given in a particular context. Even though the context is crucial for interpreting the ratings correctly, the rating-based approaches do not provide the means to represent the context explicitly. 2) The satisfaction criteria of the rater is unknown. Without knowing the expectation of the rater, it is almost impossible to make sense of a rating. We deal with these two weaknesses in two steps. First, we extend a classical rating-based approach by adding a representation of context. This addition improves the accuracy of selected service providers only when two consumers with the same service request are assumed to be satisfied with the same service. Next, we replace ratings with detailed experiences of consumers. The experiences are represented with an ontology that can capture the requested service and the received service in detail. When a service consumer decides to share her experiences with a second service consumer, the receiving consumer evaluates the experience by using her own context and satisfaction criteria. By sharing experiences rather than ratings, the service consumers can model service providers more accurately and, thus, can select service providers that are better suited for their needs.