The potential for bias and unfairness in AI-supporting government services raises ethical and legal concerns. Using crime rate prediction with the Bristol City Council data as a case study, we examine how these issues persist. Rather than auditing real-world deployed systems or producing a generalizable benchmark, our position is that a case-based investigation to understand why bias mitigations applied to government data are not always effective. Experimental analysis reveals that the failure occurs not because of flaws in model architecture or metric selection, but due to the inherent properties of the data itself, further reinforcing that the origin of bias lies in the structure and history of government datasets. We then explore the reasons for the mitigation failures in predictive models on government data and highlight the potential sources of unfairness posed by data distribution shifts, accumulated historical bias, and delays in data release. This study provides a crucial exi stence proof and serves as a critical ‘early warning’ that biases in government data may persist even with standard mitigation methods.

The evolution of Internet of Vehicles (IoV) technologies, encompassing wireless communications and Artificial Intelligence (AI), has advanced the collaborative “Pedestrian-Vehicle-Road-Cloud” IoV edge services, enhancing road efficiency and driving safety. Operating in an open-edge environment with vast sensory data, IoV faces significant privacy risks from unauthorized access and data breaches. Consequently, privacy-preserving computation (PPC) is crucial for secure IoV services. This paper reviews PPC techniques in IoV edge services, exploring network characteristics and potential privacy attacks. It categorizes and evaluates techniques such as differential privacy, homomorphic encryption, and secure multi-party computation based on data security, utility, and overhead. Summarizing their pros and cons, the challenges and future research directions for IoV edge services are outlined.

Reinforcement Learning has revolutionized decision-making processes in dynamic environments, yet it often struggles with autonomously detecting and achieving goals without clear feedback signals. For example, in a Source Term Estimation problem, the lack of precise environmental information makes it challenging to provide clear feedback signals and to define and evaluate how the source's location is determined. To address this challenge, the Autonomous Goal Detection and Cessation (AGDC) module was developed, enhancing various RL algorithms by incorporating a self-feedback mechanism for autonomous goal detection and cessation upon task completion. Our method effectively identifies and ceases undefined goals by approximating the agent's belief, significantly enhancing the capabilities of RL algorithms in environments with limited feedback. To validate effectiveness of our approach, we integrated AGDC with deep Q-Network, proximal policy optimization, and deep deterministic policy gradient algorithms, and evaluated its performance on the Source Term Estimation problem. The experimental results showed that AGDC-enhanced RL algorithms significantly outperformed traditional statistical methods such as infotaxis, entrotaxis, and dual control for exploitation and exploration, as well as a non-statistical random action selection method. These improvements were evident in terms of success rate, mean traveled distance, and search time, highlighting AGDC's effectiveness and efficiency in complex, real-world scenarios.

With the rise of social media, fake news has been known to spread rapidly through the platforms and significantly influencing public opinion. However, traditional machine learning techniques are difficult in addressing fake news detection problem due to the highly imbalanced distribution of classes. As an alternative, reinforcement learning (RL) can be used to address the imbalanced prediction problem by designing different reward functions. However, it ignores the dynamic and adversarial environment in social networks, where malicious user can influence the platform’s decision. To address these challenges, we propose a multi-agent reinforcement learning (MARL) framework called Information Classification Markov Game (ICMG) to model fake news detection as a competitive game between a malicious user and a content moderator. Within this framework, we explore the best reward function for a content moderator to achieve better performance in a real world dataset. Additionally, we compare the effectiveness of different types of agents including those based on randomness, Minimax and Q-Learning. Our results show that a MARL setting with both players using Q-Learning setting achieves the best performance. Compared with the single-agent framework, ICMG effectively improves macro F1 score, improving the model performance on the real-world dataset. Moreover, this framework effectively simulates the adversarial dynamics in real-world social media platforms, providing a new approach for developing more realistic content moderation systems.

• To identify cyber-attacks on the AI, users must have some prior understanding of the AI parameters and their normativity. • Multimodal indicators embedded across the ecosystem of AI-enabled devices are an effective way in raising users’ attention to cyber-attacks. • Engaging users to actively diagnose and resolve cyber-attacks on AI-enabled devices in the home context must take into consideration the home routines, and be designed to avoid cognitive overload. • One way to minimise overload is to make use of users’ propensity to generalise their cybersecurity knowledge and skills where possible. AI-enabled devices are increasingly introduced in the home context and cyber-attacks targeting their AI component are becoming more frequent. Moving away from seeing the user as the problem to recognising the user as part of the solution, our research reports on a novel cybersecurity intervention (comprising Explainable AI features, assisted remediation) designed to support users to identify, diagnose and mitigate cyber-attacks on the AI component of their smart devices. We carried out a case study of a bespoke smart heating device inclusive of this intervention and conducted fieldwork with ten households who experienced simulated integrity cyber-attacks over a month. Our research contributes an understanding of how to design AI-enabled devices and their ecosystems to support users to perceive integrity cyber-attacks, offering new considerations for intervention design that exploits multimodal indicators and supports users to troubleshoot themselves the causes as well as actions of cyber-attacks. Contributing to the growing area of human-centred cybersecurity, we evidence the distinctive challenges users face when evaluating integrity attacks on the AI component in the home context.

In this paper, we present the design and benchmark of an innovative sensor, ViTacTip, which fulfills the demand for advanced multi-modal sensing in a compact design. A notable feature of ViTacTip is its transparent skin, which incorporates a ‘see-through-skin’ mechanism. This mechanism aims at capturing detailed object features upon contact, significantly improving both vision-based and proximity perception capabilities. In parallel, the biomimetic tips embedded in the sensor's skin are designed to amplify contact details, thus substantially augmenting tactile and derived force perception abilities. To demonstrate the multi-modal capabilities of ViTacTip, we developed a multi-task learning model that enables simultaneous recognition of hardness, material, and textures. To assess the functionality and validate the versatility of ViTacTip, we conducted extensive benchmarking experiments, including object recognition, contact point detection, pose regression, and grating identification. To facilitate seamless switching between various sensing modalities, we employed a Generative Adversarial Network (GAN)-based approach. This method enhances the applicability of the ViTacTip sensor across diverse environments by enabling cross-modality interpretation.

Recent studies highlight the effectiveness of using in-context learning (ICL) to steer large language models (LLMs) in processing tabular data, a challenging task given the structured nature of such data.Despite advancements in performance, the fairness implications of these methods are less understood.This study investigates how varying demonstrations within ICL prompts influence the fairness outcomes of LLMs.Our findings reveal that deliberately including minority group samples in prompts significantly boosts fairness without sacrificing predictive accuracy.Further experiments demonstrate that the proportion of minority to majority samples in demonstrations affects the trade-off between fairness and prediction accuracy.Based on these insights, we introduce a mitigation technique that employs clustering and evolutionary strategies to curate a diverse and representative sample set from the training data.This approach aims to enhance both predictive performance and fairness in ICL applications.Experimental results validate that our proposed method dramatically improves fairness across various metrics, showing its efficacy in real-world scenarios.

As a popular approach in Explainable AI (XAI), an increasing number of counterfactual explanation algorithms have been proposed in the context of making machine learning classifiers more trustworthy and transparent. This paper reports our evaluations of algorithms that can output diverse counterfactuals for one instance. We first evaluate the performance of DiCE-Random, DiCE-KDTree, DiCE-Genetic and Alibi-CFRL, taking XGBoost as the machine learning model for binary classification problems. Then, we compare their suggested feature changes with feature importance by SHAP. Moreover, our study highlights that synthetic counterfactuals, drawn from the input domain but not necessarily the training data, outperform native counterfactuals from the training data regarding data privacy and validity. This research aims to guide practitioners in choosing the most suitable algorithm for generating diverse counterfactual explanations.

In this paper, we adopt constrastive explanations within an end-user application for temporal planning of smart homes. In this application, users have requirements on the execution of appliance tasks, pay for energy according to dynamic energy tariffs, have access to high-capacity battery storage, and are able to sell energy to the grid. The concurrent scheduling of devices makes this a multi-effector planning problem, while the dynamic tariffs yield costs that are non-stationary (alternatively, costs that are stationary but depend on exogenous events). These characteristics are such that the planning problems are generally not supported by existing PDDL-based planners, so we instead design a custom domain-dependent planner that scales to reasonable appliance numbers and time horizons. We conduct a controlled user study with 128 participants using an online crowd-sourcing platform based on two user stories. Our results indicate that users provided with contrastive questions and explanations have higher levels of satisfaction, tend to gain improved understanding, and rate the helpfulness more favourably with the recommended AI schedule compared to those without access to these features.

Time series forecasting, while vital in various applications, often employs complex models that are difficult for humans to understand. Effective explainable AI techniques are crucial to bridging the gap between model predictions and user understanding. This paper presents a framework - TSFeatLIME, extending TSLIME, tailored specifically for explaining univariate time series forecasting. TSFeatLIME integrates an auxiliary feature into the surrogate model and considers the pairwise Euclidean distances between the queried time series and the generated samples to improve the fidelity of the surrogate models. However, the usefulness of such explanations for human beings remains an open question. We address this by conducting a user study with 160 participants through two interactive interfaces, aiming to measure how individuals from different backgrounds can simulate or predict model output changes in the treatment group and control group. Our results show that the surrogate model under the TSFeatLIME framework is able to better simulate the behaviour of the black-box considering distance, without sacrificing accuracy. In addition, the user study suggests that the explanations were significantly more effective for participants without a computer science background.

Smart Home Heating Technologies (SHHT) have been designed to improve demand flexibility and energy conservation. SHHT rely on rational theories of energy use postulating that people will use less energy when the energy cost is higher. The inclusion of AI within SHHT is poised to optimise energy use in the future as the introduction of lower carbon energy sources place new demands on the grid. When SHHT is introduced in the home, however, they become situated in temporal heating practices that are shaped by an interplay of materiality, meanings, and competencies. We report findings from a mixed methods field study involving eleven households utilising an AI-enabled SHHT probe ‘Squid’. Taking a temporal focus throughout, our study contributes a new lens as to why households may not fully engage with SHHT's rational design, given that energy conversation is already embedded in their ongoing socio-material practices with heating. Focusing on the AI-human relation, we articulate the necessity for human agency where heating is involved, whilst also advancing an understanding of the new forms of hidden labour that households incur before they can engage with the AI. Crucially, our research informs the ongoing HCI concern over how humans understand AI, raising the question of who is responsible to assess the appropriateness of AI when the effects of human-AI performance remain opaque. Our findings contribute a new theoretical perspective into the intricate relationship between individuals and AI in the home and raise several new design implications for SHHT.

Microsurgery involves the dexterous manipulation of delicate tissue or fragile structures, such as small blood vessels and nerves, under a microscope. To address the limitations of imprecise manipulation of human hands, robotic systems have been developed to assist surgeons in performing complex microsurgical tasks with greater precision and safety. However, the steep learning curve for robot-assisted microsurgery (RAMS) and the shortage of well-trained surgeons pose significant challenges to the widespread adoption of RAMS. Therefore, the development of a versatile training system for RAMS is necessary, which can bring tangible benefits to both surgeons and patients. In this paper, we present a Tactile Internet-Based Micromanipulation System (TIMS) based on a ROS-Django web-based architecture for microsurgical training. This system can provide tactile feedback to operators via a wearable tactile display (WTD), while real-time data is transmitted through the internet via a ROS-Django framework. In addition, TIMS integrates haptic guidance to ‘guide’ the trainees to follow a desired trajectory provided by expert surgeons. Learning from demonstration based on Gaussian Process Regression (GPR) was used to generate the desired trajectory. We conducted user studies to verify the effectiveness of our proposed TIMS, comparing users' performance with and without tactile feedback and/or haptic guidance. For more details of this project, please view our website: https://sites.google.com/view/viewtims/home.

Vehicular trajectory data can be widely used in applications such as traffic prediction and congestion control. However vehicular trajectory data is voluminous and requires significant storage and processing resources, which contradicts the resources-constraint vehicular networks. Existing compression methods suffer either low compression effects or privacy leakage. A privacy-aware Trajectory Compression scheme based on Fuzzy markovian Prediction (TCFP) is proposed in this paper, which consists of two steps of fuzzy compression. The first-step compression is achieved by converting the raw trajectory data into fuzzy information on the edge vehicle sides. Further compression is performed at edge RSUs through fuzzy multi-order Markovian prediction combined with new-devised fuzzy deviation filtering rules. Extensive experimental evaluation based on real-world data sets demonstrates the proposed TCFP scheme achieves desired QoS performance in terms of compression rate, compression time and information loss.

Smart homes are powered by increasingly advanced AI, yet are controlled by, and affect, non-experts. These non-expert home users are an under-represented stakeholder in the explainable AI (XAI) literature. In this paper we facilitate future XAI research by introducing a family of smart home applications serving as a testbed to evaluate XAI with non-experts. The testbed is a hybrid-AI system spanning several AI disciplines, including machine learning and AI planning. Applications include a smart home battery and smart thermostatic radiator valve (TRV). End-user functionality is representative of leading commercial products and relevant research applications. The testbed is based on a flexible software architecture and web-based user interface, supports a range of AI tools in a modular fashion, and can be easily deployed using inexpensive consumer hardware.

There has been a sharp rise in research activities on explainable artificial intelligence (XAI), especially in the context of machine learning (ML). However, there has been less progress in developing and implementing XAI techniques in AI-enabled environments involving non-expert stakeholders. This paper reports our inves- tigations into providing explanations on the outcomes of ML algorithms to non-experts. We investigate the use of three explanation approaches (global, local, and counterfactual), considering decision trees as a use case ML model. We demonstrate the approaches with a sample dataset, and provide empirical results from a study involving over 200 participants. Our results show that most participants have a good understanding of the generated explanations.

Online social network platforms have a problem with misinformation. One popular way of addressing this problem is via the use of machine learning based automated misinformation detection systems to classify if a post is misinformation. Instead of post hoc detection, we propose to predict if a user will engage with misinformation in advance and design an effective graph neural network classifier based on ego-graphs for this task. However, social networks are highly dynamic, reflecting continual changes in user behaviour, as well as the content being posted. This is problematic for machine learning models which are typically trained on a static training dataset, and can thus become outdated when the social network changes. Inspired by the success of continual learning on such problems, we propose an ego-graphs replay strategy in continual learning (EgoCL) using graph neural networks to effectively address this issue. We have evaluated the performance of our method on user engagement with misinformation on two Twitter datasets across nineteen misinformation and conspiracy topics. Our experimental results show that our approach EgoCL has better performance in terms of predictive accuracy and computational resources than the state of the art.

Range query processing is a common edge computing and service in the Internet of things, which can extract user-interest information from distributed edge devices. How to design lightweight privacy-preserving range query processing methods remains a challenging task. Existing secure range query approaches suffer from both high communication cost and long response time, which makes them unsuitable for edge computing over resource-constrained edge devices. In this article, we propose two privacy-aware fuzzy query processing schemes based on fuzzy theory. Linguistic range variables, fuzzy overlap information, and its recovery mechanism are introduced. In addition, two distributed privacy-aware fuzzy range query processing algorithms are devised. Our approaches not only serve for privacy protection, but also aim to provide other optimal performances in terms of reliability, energy efficiency, and real-time response. Theoretical analysis and experimental evaluations based on real-world datasets validated our motivation.

Data augmentation has been widely used in machine learning for natural language processing and computer vision tasks to improve model performance. However, little research has studied data augmentation on graph neural networks, particularly using augmentation at both train- and test-time. Inspired by the success of augmentation in other domains, we have designed a method for social influence prediction using graph neural networks with train- and test-time augmentation, which can effectively generate multiple augmented graphs for social networks by utilising a variational graph autoencoder in both scenarios. We have evaluated the performance of our method on predicting user influence on multiple social network datasets. Our experimental results show that our end-to-end approach, which jointly trains a graph autoencoder and social influence behaviour classification network, can outperform state-of-the-art approaches, demonstrating the effectiveness of train-and test-time augmentation on graph neural networks for social influence prediction. We observe that this is particularly effective on smaller graphs.

Drivers always confront parking difficulties when driving on urban roads, especially in crowded downtown or beauty spots. Some of the existing literatures concentrate on multi-consideration optimization for parking decision by collecting the nearby real-time parking-related data. Others provide online parking navigation services through outsourced storage and cloud computing. Massive (raw) data transmission and complex processing are always involved in the existing methods, which results in undesired QoS such as real-time performance and privacy protection. In this paper, we propose a fuzzy skyline parking recommendation scheme for real-time parking recommendation based on roadside traffic facilities. Linguistic parking information instead of raw parking-related data is used in fuzzy skyline fusion. We evaluated our solution with real-world data sets collected from edge parking facilities in Wulin downtown, Hangzhou city, China. The evaluation results show that our approaches achieve an average accuracy of parking recommendation over 91%, low data transmission, and quick response time with privacy protection.

There has been increasing interest in using Stackelberg game (known as a security game) to allocate limited security resources against different attacker types with a specific probability distribution. However, real problems of this kind often face ambiguous information, such as imprecise, unreliable and absent payoffs, and ambiguous assignments of these payoffs. To this end, based on decision theory and the Dempster–Shafer theory of evidence, this paper proposes a novel framework that can handle these common types of ambiguity. More specifically, this paper deploys the underlying principles of existing rules from decision theory, as a way to characterise different attitudes to ambiguity, during the transformation of ambiguous payoffs into point-valued payoffs. Hence, our framework holds some good properties: (i) it subsumes traditional security games without ambiguous payoffs, (ii) a uniform margin of error will not affect the results and (iii) the influence of complete ignorance can be minimised. Also, our framework is evaluated by using nine different transformation rules, under various conditions and constraints, against 73,000 randomly generated games (a first comprehensive empirical evaluation to date). The evaluation reveals the benefits of each transformation rule and confirms that different rules can model individuals' different attitudes to ambiguity.

Recent advancements in convolutional neural networks based object detection have enabled analyzing the mounting video data with high accuracy. However, inference speed is a major drawback of these video analysis system because of the heavy object detectors. To address the computational and practicability challenges of video analysis, we propose FastQ, a system for efficient querying over video at scale. Given a target video, FastQ can automatically label the category and number of objects for each frame. We introduce a novel lightweight object detector named FDet to improve the efficiency of query system. First, a difference detector filters the frames whose difference is less than the threshold. Second, FDet is employed to efficiently label the remaining frames. To reduce inference time, FDet detects a center keypoint and a pair of corners from the feature map generated by a lightweight backbone to predict the bounding boxes. FDet completely avoid the complicated computation related to anchor boxes. Compared with state-of-the-art real-time detectors, FDet achieves superior performance with 29.1% AP on COCO benchmark at 25.3ms. Experiments show that FastQ achieves 150 times to 300 times speed-ups while maintaining more than 90% accuracy in video queries.

Efficient and accurate semantic segmentation is particularly important especially for applications like autonomous driving which requires real-time inference speed and high performance. Many works try to compromise spatial resolution to achieve real-time inference speed, which leads to poor performance. As a result, real-time segmentation task for embedded devices is still an open problem. In this paper, we focus on building a network with better performance possible while still achieve real-time inference speed. We first use a pyramid kernel size to capture more spatial information instead of using just a 3×3 kernel size for DWConvolution in ShuffleNet v2. Meanwhile, an efficient Vortex Pooling module is employed to aggregate the contextual information and generate high-resolution features. Compared with other state-of-the-art real-time semantic segmentation networks, the proposed network achieves similar inference speed and better performance on embedded device. Specifically, we achieve state-of-the-art 73.46% mean IoU on Cityscapes test dataset, for a 768×1024 input, a speed of 46.1 frames per second on NVIDIA Jetson AGX Xavier embedded development board is achieved.

The BDI architecture, where agents are modelled based on their belief, desires, and intentions, provides a practical approach to developing intelligent agents. One of the key features of BDI agents is that they are able to pursue multiple intentions in parallel, i.e. in an interleaved manner. Most of the previous works have enabled BDI agents to avoid negative interactions between intentions to ensure the correct execution. However, to avoid execution inefficiencies, BDI agents should also capitalise on positive interactions between intentions. In this paper, we provide a theoretical framework where first-principles planning (FPP) is employed to manage the intention interleaving in an automated fashion. Our FPP approach not only guarantees the achievability of intentions, but also discovers and exploits potential common sub-intentions to reduce the overall cost of intention execution. Our results show that our approach is both theoretically sound and practically feasible. The effectiveness evaluation in a manufacturing scenario shows that our approach can significantly reduce the total number of actions by merging common sub-intentions, while still accomplishing all intentions.

Closed-circuit television and sensor-based intelligent surveillance systems have attracted considerable attentions in the field of public security affairs. To provide real-time reaction in the case of a huge volume of the surveillance data, researchers have proposed event-reasoning frameworks for modeling and inferring events of interest. However, they do not support decision-making, which is very important for surveillance operators. To this end, this paper incorporate a function of decision-making in an event-reasoning framework, so that our model not only can perform event-reasoning but also can predict, rank, and alarm threats according to uncertain information from multiple heterogeneous sources. In particular, we propose a multiattribute decision-making model, in which an object being watched is modeled as a multiattribute event, where each attribute corresponds to a specific source, and the information from each source can be used to elicit a local threat degree of different malicious situations with respect to the corresponding attribute. Moreover, to assess an overall threat degree of an object being observed, we also propose a method to fuse the conflict threat degrees regarding all the relevant attributes. Finally, we demonstrate the effectiveness of our framework by an airport security surveillance scenario.

Some well-known paradoxes in decision making (e.g., the Allais paradox, the St. Peterburg paradox, the Ellsberg paradox, and the Machina paradox) reveal that choices conventional expected utility theory predicts could be inconsistent with empirical observations. So, solutions to these paradoxes can help us better understand humans decision making accurately. This is also highly related to the prediction power of a decision-making model in real-world applications. Thus, various models have been proposed to address these paradoxes. However, most of them can only solve parts of the paradoxes, and for doing so some of them have to rely on the parameter tuning without proper justifications for such bounds of parameters. To this end, this paper proposes a new descriptive decision-making model, expected utility with relative loss reduction, which can exhibit the same qualitative behaviours as those observed in experiments of these paradoxes without any additional parameter setting. In particular, we introduce the concept of relative loss reduction to reflect people's tendency to prefer ensuring a sufficient minimum loss to just a maximum expected utility in decision-making under risk or ambiguity.

The Belief-Desire-Intention (BDI) paradigm is a flexible framework for representing intelligent agents. Practical BDI agent systems rely on a static plan library to reduce the planning problem to the simpler problem of plan selection. However, fixed pre-defined plan libraries are unable to adapt to fast-changing environments pervaded by uncertainty. In this paper, we advance the state-of-the-art in BDI agent systems by proposing a plan library evolution architecture with mechanisms to incorporate new plans (plan expansion) and drop old/unsuitable plans (plan contraction) to adapt to changes in a realistic environment. The proposal follows a principled approach to define plan library expansion and contraction operators, motivated by postulates that clearly highlight the underlying assumptions, and quantified by decision-support measures of temporal information. In particular, we demonstrate the feasibility of the proposed contraction operator by presenting a multi-criteria argumentation based decision making to remove plans exemplified in a planetary vehicle scenario.

This paper presents an event modelling and reasoning framework where event-observations obtained from heterogeneous sources may be uncertain or incomplete, while sensors may be unreliable or in conflict. To address these issues we apply Dempster-Shafer (DS) theory to correctly model the event-observations so that they can be combined in a consistent way. Unfortunately, existing frameworks do not specify which event-observations should be selected to combine. Our framework provides a rule-based approach to ensure combination occurs on event-observations from multiple sources corresponding to the same event of an individual subject. In addition, our framework provides an inference rule set to infer higher level inferred events by reasoning over the uncertain event-observations as epistemic states using a formal language. Finally, we illustrate the usefulness of the framework using a sensor-based surveillance scenario.

Factored MDPs provide an efficient way to reduce the complexity of large, real-world domains by exploiting structure within the state space. This avoids the need for the state space to be fully enumerated, which is impractical in large domains. However, defining a reward function for state transitions is difficult in a factored MDP since transitions are not known prior to execution. In this paper, we provide a novel method for deriving rewards from information within the states in order to determine intermediate rewards for state transitions. We do this by treating some specific state variables as resources, allowing costs and rewards to be inferred from changes to the resources and ensuring the agent is resource-aware while also being goal-oriented. To facilitate this, we propose a novel variant of Dynamic Bayesian Networks specifically for modelling action transitions and capable of dealing with relative changes to realvalued state variables (such as resources) in a compact fashion. We also propose a number of reward functions which model resource types commonly found in real-world situations. We go on to show that our proposed framework offers an improvement over existing techniques involving reward functions for factored MDPs as it improves both the efficiency and decision quality of online planners when operating on these models.

The Belief-Desire-Intention (BDI) architecture is a practical approach for modelling large-scale intelligent systems. In the BDI setting, a complex system is represented as a network of interacting agents - or components - each one modelled based on its beliefs, desires and intentions. However, current BDI implementations are not well-suited for modelling more realistic intelligent systems which operate in environments pervaded by different types of uncertainty. Furthermore, existing approaches for dealing with uncertainty typically do not offer syntactical or tractable ways of reasoning about uncertainty. This complicates their integration with BDI implementations, which heavily rely on fast and reactive decisions. In this paper, we advance the state-of-the-art w.r.t. handling different types of uncertainty in BDI agents. The contributions of this paper are, first, a new way of modelling the beliefs of an agent as a set of epistemic states. Each epistemic state can use a distinct underlying uncertainty theory and revision strategy, and commensurability between epistemic states is achieved through a stratification approach. Second, we present a novel syntactic approach to revising beliefs given unreliable input. We prove that this syntactic approach agrees with the semantic definition, and we identify expressive fragments that are particularly useful for resource-bounded agents. Third, we introduce full operational semantics that extend CAN, a popular semantics for BDI, to establish how reasoning about uncertainty can be tightly integrated into the BDI framework. Fourth, we provide comprehensive experimental results to highlight the usefulness and feasibility of our approach, and explain how the generic epistemic state can be instantiated into various representations.

There has been much interest in the belief–desire–intention (BDI) agent-based model for developing scalable intelligent systems, e.g. using the AgentSpeak framework. However, reasoning from sensor information in these large-scale systems remains a significant challenge. For example, agents may be faced with information from heterogeneous sources which is uncertain and incomplete, while the sources themselves may be unreliable or conflicting. In order to derive meaningful conclusions, it is important that such information be correctly modelled and combined. In this paper, we choose to model uncertain sensor information in Dempster–Shafer (DS) theory. Unfortunately, as in other uncertainty theories, simple combination strategies in DS theory are often too restrictive (losing valuable information) or too permissive (resulting in ignorance). For this reason, we investigate how a context-dependent strategy originally defined for possibility theory can be adapted to DS theory. In particular, we use the notion of largely partially maximal consistent subsets (LPMCSes) to characterise the context for when to use Dempster’s original rule of combination and for when to resort to an alternative. To guide this process, we identify existing measures of similarity and conflict for finding LPMCSes along with quality of information heuristics to ensure that LPMCSes are formed around high-quality information. We then propose an intelligent sensor model for integrating this information into the AgentSpeak framework which is responsible for applying evidence propagation to construct compatible information, for performing context-dependent combination and for deriving beliefs for revising an agent’s belief base. Finally, we present a power grid scenario inspired by a real-world case study to demonstrate our work.

The ability of an autonomous agent to select rational actions is vital in enabling it to achieve its goals. To do so effectively in a high-stakes setting, the agent must be capable of considering the risk and potential reward of both immediate and future actions. In this paper we provide a novel method for calculating risk alongside utility in online planning algorithms. We integrate such a risk-aware planner with a BDI agent, allowing us to build agents that can set their risk aversion levels dynamically based on their changing beliefs about the environment. To guide the design of a risk-aware agent we propose a number of principles which such an agent should adhere to and show how our proposed framework satisfies these principles. Finally, we evaluate our approach and demonstrate that a dynamically risk-averse agent is capable of achieving a higher success rate than an agent that ignores risk, while obtaining a higher utility than an agent with a static risk attitude.

Most traditional data mining algorithms struggle to cope with the sheer scale of data efficiently. In this paper, we propose a general framework to accelerate existing algorithms to cluster large-scale datasets which contain large numbers of attributes, items, and clusters. Our framework makes use of locality sensitive hashing to significantly reduce the cluster search space. We also theoretically prove that our framework has a guaranteed error bound in terms of the clustering quality. This framework can be applied to a set of centroid-based clustering algorithms that assign an object to the most similar cluster, and we adopt the popular K-Modes categorical clustering algorithm to present how the framework can be applied. We validated our framework with five synthetic datasets and a real world Yahoo! Answers dataset. The experimental results demonstrate that our framework is able to speed up the existing clustering algorithm between factors of 2 and 6, while maintaining comparable cluster purity.

Planning is an essential process in teams of multiple agents pursuing a common goal. When the effects of actions undertaken by agents are uncertain, evaluating the potential risk of such actions alongside their utility might lead to more rational decisions upon planning. This challenge has been recently tackled for single agent settings, yet domains with multiple agents that present diverse viewpoints towards risk still necessitate comprehensive decision making mechanisms that balance the utility and risk of actions. In this work, we propose a novel collaborative multi-agent planning framework that integrates (i) a team-level online planner under uncertainty that extends the classical UCT approximate algorithm, and (ii) a preference modeling and multicriteria group decision making approach that allows agents to find accepted and rational solutions for planning problems, predicated on the attitude each agent adopts towards risk. When utilised in risk-pervaded scenarios, the proposed framework can reduce the cost of reaching the common goal sought and increase effectiveness, before making collective decisions by appropriately balancing risk and utility of actions.

Recommending users for a new social network user to follow is a topic of interest at present. The existing approaches rely on using various types of information about the new user to determine recommended users who have similar interests to the new user. However, this presents a problem when a new user joins a social network, who is yet to have any interaction on the social network. In this paper we present a particular type of conversational recommendation approach, critiquing-based recommendation, to solve the cold start problem. We present a critiquing-based recommendation system, called CSFinder, to recommend users for a new user to follow. A traditional critiquing-based recommendation system allows a user to critique a feature of a recommended item at a time and gradually leads the user to the target recommendation. However this may require a lengthy recommendation session. CSFinder aims to reduce the session length by taking a case-based reasoning approach. It selects relevant recommendation sessions of past users that match the recommendation session of the current user to short-cut the current recommendation session. It selects relevant recommendation sessions from a case base that contains the successful recommendation sessions of past users. A past recommendation session can be selected if it contains recommended items and critiques that sufficiently overlap with the ones in the current session. Our experimental results show that CSFinder has significantly shorter sessions than the ones of an Incremental Critiquing system, which is a baseline critiquing-based recommendation system.

Necessary and sufficient conditions for choice functions to be rational have been intensively studied in the past. However, in these attempts, a choice function is completely specified. That is, given any subset of options, called an issue, the best option over that issue is always known, whilst in real-world scenarios, it is very often that only a few choices are known instead of all. In this paper, we study partial choice functions and investigate necessary and sufficient rationality conditions for situations where only a few choices are known. We prove that our necessary and sufficient condition for partial choice functions boils down to the necessary and sufficient conditions for complete choice functions proposed in the literature. Choice functions have been instrumental in belief revision theory. That is, in most approaches to belief revision, the problem studied can simply be described as the choice of possible worlds compatible with the input information, given an agent’s prior belief state. The main effort has been to devise strategies in order to infer the agents revised belief state. Our study considers the converse problem: given a collection of input information items and their corresponding revision results (as provided by an agent), does there exist a rational revision operation used by the agent and a consistent belief state that may explain the observed results?

In this paper we propose a graph stream clustering algorithm with a unified similarity measure on both structural and attribute properties of vertices, with each attribute being treated as a vertex. Unlike others, our approach does not require an input parameter for the number of clusters, instead, it dynamically creates new sketch-based clusters and periodically merges existing similar clusters. Experiments on two publicly available datasets reveal the advantages of our approach in detecting vertex clusters in the graph stream. We provide a detailed investigation into how parameters affect the algorithm performance. We also provide a quantitative evaluation and comparison with a well-known offline community detection algorithm which shows that our streaming algorithm can achieve comparable or better average cluster purity.

This paper presents an event recognition framework, based on Dempster-Shafer theory, that combines evidence of events from low-level computer vision analytics. The proposed method employing evidential network modelling of composite events, is able to represent uncertainty of event output from low level video analysis and infer high-level events with semantic meaning along with degrees of belief. The method has been evaluated on videos taken of subjects entering and leaving a seated area. This has relevance to a number of transport scenarios, such as onboard buses and trains, and also in train stations and airports. Recognition results of 78% and 100% for four composite events are encouraging.

Three issues usually are associated with threat prevention intelligent surveillance systems. First, the fusion and interpretation of large scale incomplete heterogeneous information; second, the demand of effectively predicting suspects' intention and ranking the potential threats posed by each suspect; third, strategies of allocating limited security resources (e.g., the dispatch of security team) to prevent a suspect's further actions towards critical assets. However, in the literature, these three issues are seldomly considered together in a sensor network based intelligent surveillance framework. To address this problem, in this paper, we propose a multi-level decision support framework for in-time reaction in intelligent surveillance. More specifically, based on a multi-criteria event modeling framework, we design a method to predict the most plausible intention of a suspect. Following this, a decision support model is proposed to rank each suspect based on their threat severity and to determine resource allocation strategies. Finally, formal properties are discussed to justify our framework.

Revising its beliefs when receiving new information is an important ability of any intelligent system. However, in realistic settings the new input is not always certain. A compelling way of dealing with uncertain input in an agent-based setting is to treat it as unreliable input, which may strengthen or weaken the beliefs of the agent. Recent work focused on the postulates associated with this form of belief change and on finding semantical operators that satisfy these postulates. In this paper we propose a new syntactic approach for this form of belief change and show that it agrees with the semantical definition. This makes it feasible to develop complex agent systems capable of efficiently dealing with unreliable input in a semantically meaningful way. Additionally, we show that imposing restrictions on the input and the beliefs that are entailed allows us to devise a tractable approach suitable for resource-bounded agents or agents where reactive ness is of paramount importance.

When an agent wants to fulfill its desires about the world, the agent usually has multiple plans to choose from and these plans have different pre-conditions and additional effects in addition to achieving its goals. Therefore, for further reasoning and interaction with the world, a plan selection strategy (usually based on plan cost estimation) is mandatory for an autonomous agent. This demand becomes even more critical when uncertainty on the observation of the world is taken into account, since in this case, we consider not only the costs of different plans, but also their chances of success estimated according to the agent's beliefs. In addition, when multiple goals are considered together, different plans achieving the goals can be conflicting on their preconditions (contexts) or the required resources. Hence a plan selection strategy should be able to choose a subset of plans that fulfills the maximum number of goals while maintaining context consistency and resource-tolerance among the chosen plans. To address the above two issues, in this paper we first propose several principles that a plan selection strategy should satisfy, and then we present selection strategies that stem from the principles, depending on whether a plan cost is taken into account. In addition, we also show that our selection strategy can partially recover intention revision.

This paper presents a multi-agent system approach to address the difficulties encountered in traditional SCADA systems deployed in critical environments such as electrical power generation, transmission and distribution.The approach models uncertainty and combines multiple sources of uncertain information to deliver robust plan selection.We examine the approach in the context of a simplified power supply/demand scenario using a residential grid connected solar system and consider the challenges of modelling and reasoning with uncertain sensor information in this environment.We discuss examples of plans and actions required for sensing, establish and discuss the effect of uncertainty on such systems and investigate different uncertainty theories and how they can fuse uncertain information from multiple sources for effective decision making in such a complex system.

Decision makers are required to make critical decisions throughout all stages of a life-cycle in large-scale projects. These decisions are important as they impact upon the outcome and the success of projects. In this paper we present an evidential reasoning framework to aid decision-makers in the decision making process. This approach utilizes the Dezert-Smarandache Theory (DSm) to fuse heterogeneous evidence sources that suffer from levels of uncertainty, imprecision and conflicts to provide beliefs for decision options. To analyze the impact that source reliability and priority has upon the decision making process a reliability discounting technique along with a priority discounting technique are applied. Application of the evidential reasoning framework is illustrated using a Case Study based in the Aerospace domain.

It is increasingly recognized that identifying the degree of blame or responsibility of each formula for inconsistency of a knowledge base (i.e. a set of formulas) is useful for making rational decisions to resolve inconsistency in that knowledge base. Most current techniques for measuring the blame of each formula with regard to an inconsistent knowledge base focus on classical knowledge bases only. Proposals for measuring the blames of formulas with regard to an inconsistent prioritized knowledge base have not yet been given much consideration. However, the notion of priority is important in inconsistency-tolerant reasoning. This article investigates this issue and presents a family of measurements for the degree of blame of each formula in an inconsistent prioritized knowledge base by using the minimal inconsistent subsets of that knowledge base. First of all, we present a set of intuitive postulates as general criteria to characterize rational measurements for the blames of formulas of an inconsistent prioritized knowledge base. Then we present a family of measurements for the blame of each formula in an inconsistent prioritized knowledge base under the guidance of the principle of proportionality, one of the intuitive postulates. We also demonstrate that each of these measurements possesses the properties that it ought to have. Finally, we use a simple but explanatory example in requirements engineering to illustrate the application of these measurements. Compared to the related works, the postulates presented in this article consider the special characteristics of minimal inconsistent subsets as well as the priority levels of formulas. This makes them more appropriate to characterizing the inconsistency measures defined from minimal inconsistent subsets for prioritized knowledge bases as well as classical knowledge bases. Correspondingly, the measures guided by these postulates can intuitively capture the inconsistency for prioritized knowledge bases.

This paper presents Yagada, an algorithm to search labelled graphs for anomalies using both structural data and numeric attributes. Yagada is explained using several security-related examples and validated with experiments on a physical Access Control database. Quantitative analysis shows that in the upper range of anomaly thresholds, Yagada detects twice as many anomalies as the best-performing numeric discretization algorithm. Qualitative evaluation shows that the detected anomalies are meaningful, representing a combination of structural irregularities and numerical outliers.

The number of clinical trials reports is increasing rapidly due to a large number of clinical trials being conducted; it, therefore, raises an urgent need to utilize the clinical knowledge contained in the clinical trials reports. In this paper, we focus on the qualitative knowledge instead of quantitative knowledge. More precisely, we aim to model and reason with the qualitative comparison (QC for short) relations which consider qualitatively how strongly one drug/therapy is preferred to another in a clinical point of view. To this end, first, we formalize the QC relations, introduce the notions of QC language, QC base, and QC profile; second, we propose a set of induction rules for the QC relations and provide grading interpretations for the QC bases and show how to determine whether a QC base is consistent. Furthermore, when a QC base is inconsistent, we analyze how to measure inconsistencies among QC bases, and we propose different approaches to merging multiple QC bases. Finally, a case study on lowering intraocular pressure is conducted to illustrate our approaches. © 2010 Wiley Periodicals, Inc.

Belief revision performs belief change on an agent's beliefs when new evidence (either of the form of a propositional formula or of the form of a total pre-order on a set of interpretations) is received. Jeffrey's rule is commonly used for revising probabilistic epistemic states when new information is probabilistically uncertain. In this paper, we propose a general epistemic revision framework where new evidence is of the form of a partial epistemic state. Our framework extends Jeffrey's rule with uncertain inputs and covers well-known existing frameworks such as ordinal conditional function (OCF) or possibility theory. We then define a set of postulates that such revision operators shall satisfy and establish representation theorems to characterize those postulates. We show that these postulates reveal common characteristics of various existing revision strategies and are satisfied by OCF conditionalization, Jeffrey's rule of conditioning and possibility conditionalization. Furthermore, when reducing to the belief revision situation, our postulates can induce most of Darwiche and Pearl's postulates.

When merging belief sets from different agents, the result is normally a consistent belief set in which the inconsistency between the original sources is not represented. As probability theory is widely used to represent uncertainty, an interesting question therefore is whether it is possible to induce a probability distribution when merging belief sets. To this end, we first propose two approaches to inducing a probability distribution on a set of possible worlds, by extending the principle of indifference on possible worlds. We then study how the (in)dependence relations between atoms can influence the probability distribution. We also propose a set of properties to regulate the merging of belief sets when a probability distribution is output. Furthermore, our merging operators satisfy the well known Konieczny and Pino-Perez postulates if we use the set of possible worlds which have the maximal induced probability values. Our study shows that taking an induced probability distribution as a merging result can better reflect uncertainty and inconsistency among the original knowledge bases.

The Intelligent Sensor Information System (ISIS) is described. ISIS is an active CCTV approach to reducing crime and anti-social behavior on public transport systems such as buses. Key to the system is the idea of event composition, in which directly detected atomic events are combined to infer higher-level events with semantic meaning. Video analytics are described that profile the gender of passengers and track them as they move about a 3-D space. The overall system architecture is described which integrates the on-board event recognition with the control room software over a wireless network to generate a real-time alert. Data from preliminary data-gathering trial is presented.

Combination rules proposed so far in the Dempster-Shafer theory of evidence, especially Dempster rule, rely on a basic assumption, that is, pieces of evidence being combined are considered to be on a par, i.e. play the same role. When a source of evidence is less reliable than another, it is possible to discount it and then a symmetric combination operation is still used. In the case of revision, the idea is to let prior knowledge of an agent be altered by some input information. The change problem is thus intrinsically asymmetric. Assuming the input information is reliable, it should be retained whilst the prior information should be changed minimally to that effect. Although belief revision is already an important subfield of artificial intelligence, so far, it has been little addressed in evidence theory. In this paper, we define the notion of revision for the theory of evidence and propose several different revision rules, called the inner and outer revisions, and a modified adaptive outer revision, which better corresponds to the idea of revision. Properties of these revision rules are also investigated.

Demand for bus surveillance is growing due to the increased threats of terrorist attack, vandalism and litigation. However, CCTV systems are traditionally used in forensic mode, precluding an in-time reaction to an event. In this paper, we introduce a real-time event composition framework which can support the instant recognition of emergent events based on uncertain or imperfect information gathered from multiple sources. This framework deploys a rule-based reasoning component that can infer malicious situations (composite events) from a set of correlated atomic events. These are recognized by applying analytic algorithms to the multimedia contents of bus surveillance data. We demonstrate the significance and usefulness of our framework with a case study of an on-going bus surveillance project.

Developing a desirable framework for handling inconsistencies in software requirements specifications is a challenging problem. It has been widely recognized that the relative priority of requirements can help developers to make some necessary trade-off decisions for resolving con- flicts. However, for most distributed development such as viewpoints-based approaches, different stakeholders may assign different levels of priority to the same shared requirements statement from their own perspectives. The disagreement in the local levels of priority assigned to the same shared requirements statement often puts developers into a dilemma during the inconsistency handling process. The main contribution of this paper is to present a prioritized merging-based framework for handling inconsistency in distributed software requirements specifications. Given a set of distributed inconsistent requirements collections with the local prioritization, we first construct a requirements specification with a prioritization from an overall perspective. We provide two approaches to constructing a requirements specification with the global prioritization, including a merging-based construction and a priority vector-based construction. Following this, we derive proposals for handling inconsistencies from the globally prioritized requirements specification in terms of prioritized merging. Moreover, from the overall perspective, these proposals may be viewed as the most appropriate to modifying the given inconsistent requirements specification in the sense of the ordering relation over all the consistent subsets of the requirements specification. Finally, we consider applying negotiation-based techniques to viewpoints so as to identify an acceptable common proposal from these proposals.

In this paper, we present a general revision model on epistemic states based on plausibility measures proposed by Friedman and Halpern. We propose our revision strategy and give some desirable properties, e.g., the reversible and commutative properties. Moreover, we develop a notion called plausibility kinematics and show that our revision strategy follows plausibility kinematics. Furthermore, we prove that the revision following plausibility kinematics satisfies the principle of minimal change based on some distance measures. Finally, we discuss a revision operator defined for plausibility functions and its relationship with iterated belief revision proposed by Darwiche and Pearl. We show that the revision operator satisfies all the DP postulates when it is Max-Additive.

In this paper, we present a revision strategy of revising a conditional probabilistic logic program (PLP) when new information is received (which is in the form of probabilistic formulae), through the technique of variable forgetting. We first extend the traditional forgetting method to forget a conditional event in PLPs. We then propose two revision operators to revise a PLP based on our forgetting method. By revision through forgetting, the irrelevant knowledge in the original PLP is retained according to the minimal change principle. We prove that our revision operators satisfy most of the postulates for probabilistic belief revision. A main advantage of our revision operators is that a new PLP is explicitly obtained after revision, since our revision operator performs forgetting a conditional event at the syntax level.

The need to merge multiple sources of uncertain information is an important issue in many application areas, particularly when there is potential for contradictions between sources. Possibility theory offers a flexible framework to represent, and reason with, uncertain information, and there is a range of merging operators, such as the conjunctive and disjunctive operators, for combining information. However, with the proposals to date, the context of the information to be merged is largely ignored during the process of selecting which merging operators to use. To address this shortcoming, in this paper, we propose an adaptive merging algorithm which selects largely partially maximal consistent subsets of sources, which can be merged through the relaxation of the conjunctive operator, by assessing the coherence of the information in each subset. In this way, a fusion process can integrate both conjunctive and disjunctive operators in a more flexible manner and thereby be more context dependent. A comparison with related merging methods shows how our algorithm can produce a more consensual result.

The study of alternative combination rules in DS theory when evidence is in conflict has emerged again recently as an interesting topic, especially in data/information fusion applications. These studies have mainly focused on investigating which alternative would be appropriate for which conflicting situation, under the assumption that a conflict is identified. The issue of detection (or identification) of conflict among evidence has been ignored. In this paper, we formally define when two basic belief assignments are in conflict. This definition deploys quantitative measures of both the mass of the combined belief assigned to the emptyset before normalization and the distance between betting commitments of beliefs. We argue that only when both measures are high, it is safe to say the evidence is in conflict. This definition can be served as a prerequisite for selecting appropriate combination rules.

In the environment of data integration over the Internet, three major factors affect the cost of a query: network congestion situation, server contention states (workload), and data/query complexity. We concentrate on system contention states. For a remote data source, we first determine the total number of contention states of the system through applying clustering techniques to the costs of sample queries. We then develop a set of cost formulae for each of the contention states using a multiple regression process. Finally, we estimate the system's current contention state when a query is issued and using either a time slides method or a statistical method depending on the information we have about the system. Our method can accurately predict the system contention state so that the effect of the contention states on the cost of queries can be estimated precisely.

This paper presents a method of partitioning a Markov tree of belief functions into clusters so as to efficiently implement parallel belief function propagations on the basis of the local computation technique. Our method initially represents computations of combining evidence on all nodes in a Markov tree as parallelism instances, then balances the computation load among these instances, and finally partitions them into clusters which can be mapped onto a set of processors in a PowerPC network. The advantage of our method is that the maximum parallelization can still be achieved, even with limited processor availability.

This paper presents an efficient algorithm for learning Bayesian belief networks from databases.The algorithm takes a database as input and constructs the belief network structure as output.The construction process is based on the computation of mutual information of attribute pairs.Given a data set that is large enough, this algorithm can generate a belief network very close to the underlying model, and at the same time, enjoys the time complexity of 0(N4) on conditional independence (CI) tests.When the data set has a normal DAG-Faithfil (see Section 3.2) probability distribution, the algorithm guarantees that the structure of a perfect map pearl, 19881 of the underlying dependency mode1 is generated.To evaluate this algorithm, we present the experimental results on three versions of the wellknown ALARM nehvork database, which has 37 attributes and 10,000 records.The results show that this algorithm is accurate and efficient.The proof of correctness and the analysis of computational complexity are also presented. 1 -*. 1 .

Describes a practicable approach to extend relational databases to deductive databases by using a two-level knowledge representation mechanism. By this approach, we developed a deductive database add-in module that can be easily integrated with relational databases. Therefore, users can get deductive reasoning functionality without changing their favourite relational databases and query languages. This approach proved to be efficient on knowledge base organization and reasoning in real-world applications.

Use of the Dempster-Shafer (D-S) theory of evidence to deal with uncertainty in knowledge-based systems has been widely addressed. Several AI implementations have been undertaken based on the D-S theory of evidence or the extended theory. But the representation of uncertain relationships between evidence and hypothesis groups (heuristic knowledge) is still a major problem. This paper presents an approach to representing such knowledge, in which Yen’s probabilistic multi-set mappings have been extended to evidential mappings, and Shafer’s partition technique is used to get the mass function in a complex evidence space. Then, a new graphic method for describing the knowledge is introduced which is an extension of the graphic model by Lowrance et al. Finally, an extended framework for evidential reasoning systems is specified.