Abstract Indirect reciprocity, supported by simple reputation assessment and social norms, has been demonstrated as an effective mechanism for enabling cooperation in populations of self-interested individuals. However, it has been shown that where there is noise in the performance of actions, or in observers’ perceptions, cooperation may not emerge. Higher-order social norms and generosity have been investigated as potential mechanisms to support cooperation in such environments, but are ineffective without additional, limiting, assumptions. In particular, higher-order norms have typically been investigated in cases where reputation is binary (‘good’ or ‘bad’) and where all agents ascribe the same reputation to an individual, implying full and perfect observation of actions. Generosity with an ‘aligned’ strategy, where all agents have the same likelihood of being generous, has been shown to be ineffective where reputation is binary. In this work we consider reciprocity emergence mechanisms in noisy domains and where agent observations may be incomplete or inaccurate. Our hypothesis is that nuanced reputation scores will enable generosity to be more effective, since an individual act of generosity will have a less extreme impact on reputation. We also investigate whether replacing the ‘aligned’ setting for generosity with a ‘non-aligned’ alternative, which we refer to as forgiveness, will support cooperation in noisy partially observable environments, without the level of ‘unjustified benevolence’ exhibited by generosity. We show both analytically and empirically that generosity when combined with fine grained reputation can help cooperation emerge, and that forgiveness can support cooperation in certain settings.

Research has shown that cooperative action struggles to emerge in the noisy variant of the donation game, a simple model of noisy multi-agent systems where indirect reciprocity is required to maximise utility. Such noise can arise when agents may have an incorrect view of the reputation of their interaction partners, or when the actions themselves may fail. Concepts such as generosity, as well as the use of higher-order norms, have been investigated as mechanisms to facilitate cooperation in such environments, but often are not effective or require additional assumptions or infrastructure in the system to operate. In this paper, we demonstrate both analytically and empirically that a simple form of generosity when combined with fine grained reputation can help cooperation emerge. We also show that the use of individual forgiveness strategies rather than the presence of global generosity can support cooperation in such environments.

Abstract Autonomous agents are becoming increasingly ubiquitous and are playing an increasing role in wide range of safety-critical systems, such as driverless cars, exploration robots and unmanned aerial vehicles. These agents operate in highly dynamic and heterogeneous environments, resulting in complex behaviour and interactions. Therefore, the need arises to model and understand more complex and nuanced agent interactions than have previously been studied. In this paper, we propose a novel agent-based modelling approach to investigating norm emergence, in which such interactions can be investigated. To this end, while there may be an ideal set of optimally compatible actions there are also combinations that have positive rewards and are also compatible. Our approach provides a step towards identifying the conditions under which globally compatible norms are likely to emerge in the context of complex rewards. Our model is illustrated using the motivating example of self-driving cars, and we present the scenario of an autonomous vehicle performing a left-turn at a T-intersection.

Given the distributed, heterogenous, and dynamic nature of service-based IoT systems, capturing circumstances data underlying service provisions becomes increasingly important for understanding process flow and tracing how outputs came about, thus enabling clients to make more informed decisions regarding future interaction partners. Whilst service providers are the main source of such circumstances data, they may often be reluctant to release it, e.g., due to the cost and effort required, or to protect their interests. In response, this article introduces a reputation-based framework, guided by intelligent software agents, to support the sharing of truthful circumstances information by providers. In this framework, assessor agents , acting on behalf of clients, rank and select service providers according to reputation, while provider agents , acting on behalf of service providers, learn from the environment and adjust provider’s circumstances provision policies in the direction that increases provider profit with respect to perceived reputation. The novelty of the reputation assessment model adopted by assessor agents lies in affecting provider reputation scores by whether or not they reveal truthful circumstances data underlying their service provisions, in addition to other factors commonly adopted by existing reputation schemes. The effectiveness of the proposed framework is demonstrated through an agent-based simulation including robustness against a number of attacks, with a comparative performance analysis against FIRE as a baseline reputation model.

The rise of intelligent vehicle systems will lead to more human-machine interactions and so there is a need to create a bridge between the system and the actions and behaviours of the people inside the vehicle. In this paper, we propose a dual camera setup to monitor the actions and behaviour of vehicle passengers and a deep learning architecture which can utilise video data to classify a range of actions. The method incorporates two different views as input to a 3D convolutional network and uses transfer learning from other action recognition data. The performance of this method is evaluated using an in-vehicle dataset, which contains video recordings of people performing a range of common in-vehicle actions. We show that the combination of transfer learning and using dual viewpoints in a 3D action recognition network offers an increase in classification accuracy of action classes with distinct poses, e.g. mobile phone use and sleeping, whilst it does not apply as well for classifying those actions with small movements, such as talking and eating.

Abstract Quality of Service (QoS) properties play an important role in distinguishing between functionally equivalent services and accommodating the different expectations of users. However, the subjective nature of some properties and the dynamic and unreliable nature of service environments may result in cases where the quality values advertised by the service provider are either missing or untrustworthy. To tackle this, a number of QoS estimation approaches have been proposed, using the observation history available on a service to predict its performance. Although the context underlying such previous observations (and corresponding to both user and service related factors) could provide an important source of information for the QoS estimation process, it has only been used to a limited extent by existing approaches. In response, we propose a context‐aware quality learning model, realized via a learning‐enabled service agent, exploiting the contextual characteristics of the domain to provide more personalized, accurate, and relevant quality estimations for the situation at hand. The experiments conducted demonstrate the effectiveness of the proposed approach, showing promising results (in terms of prediction accuracy) in different types of changing service environments.

Norms are a valuable means of establishing coherent cooperative behaviour in decentralised systems in which there is no central authority. Axelrod’s seminal model of norm establishment in populations of self-interested individuals provides some insight into the mechanisms needed to support this through the use of metanorms, but considers only limited scenarios and domains. While further developments of Axelrod’s model have addressed some of the limitations, there is still only limited consideration of such metanorm models with more realistic topological configurations. In response, this paper tries to address such limitation by considering its application to different topological structures. Our results suggest that norm establishment is achievable in lattices and small worlds, while such establishment is not achievable in scale-free networks, due to the problematic effects of hubs. The paper offers a solution, first by adjusting the model to more appropriately reflect the characteristics of the problem, and second by offering a new dynamic policy adaptation approach to learning the right behaviour. Experimental results demonstrate that this dynamic policy adaptation overcomes the difficulties posed by the asymmetric distribution of links in scale-free networks, leading to an absence of norm violation, and instead to norm emergence.

Predicting the future actions of individuals from geospatial data has the potential to provide a basis for tailored services. This work presents the Predictive Context Tree (PCT), a new hierarchical classifier based on the Context Tree summary model [8]. The PCT is capable of predicting the future contexts and locations of individuals to provide a basis for understanding not only where a user will be, but also what type of activity they will be performing. Through a comparison to established techniques, this paper demonstrates the applicability of the PCT by showing increased accuracies for location prediction, and increased utility through context prediction.

Exposing latent knowledge in geospatial trajectories has the potential to provide a better understanding of the movements of individuals and groups. Motivated by such a desire, this work presents the context tree , a new hierarchical data structure that summarises the context behind user actions in a single model. We propose a method for context tree construction that augments geospatial trajectories with land usage data to identify such contexts. Through evaluation of the construction method and analysis of the properties of generated context trees, we demonstrate the foundation for understanding and modelling behaviour afforded. Summarising user contexts into a single data structure gives easy access to information that would otherwise remain latent, providing the basis for better understanding and predicting the actions and behaviours of individuals and groups. Finally, we also present a method for pruning context trees for use in applications where it is desirable to reduce the size of the tree while retaining useful information.

Driving is a safety critical task that requires a high levels of attention and workload from the driver. Despite this, people often also perform secondary tasks such as eating or using a mobile phone, which increase workload levels and divert cognitive and physical attention from the primary task of driving. If a vehicle is aware that the driver is currently under high workload, the vehicle functionality can be changed in order to minimize any further demand. Traditionally, workload measurements have been performed using intrusive means such as physiological sensors. Another approach may be to monitor workload online using readily available and robust sensors accessible via the vehicle's Controller Area Network (CAN). In this paper, we present details of the Warwick-JLR Driver Monitoring Dataset (DMD) collected for this purpose, and to announce its publication for driver monitoring research. The collection protocol is briefly introduced, followed by statistical analysis of the dataset to describe its structure. Finally, the public release of the dataset, for use in both driver monitoring and data mining research, is announced.

Establishing cooperation and protecting individuals from selfish and malicious behavior are key goals in open multiagent systems. Incomplete information regarding potential interaction partners can undermine typical cooperation mechanisms such as trust and reputation, particularly in lightweight systems designed for individuals with significant resource constraints. In this article, we (i) propose extending a low‐cost reputation mechanism to use gossiping to mitigate against the effect of incomplete information, (ii) define four simple aggregation strategies for incorporating gossiped information, and (iii) evaluate our model on a variety of synthetic and real‐world topologies and under a range of configurations. We show that (i) gossiping can significantly reduce the potentially detrimental influence of incomplete information and the underlying network structure on lightweight reputation mechanisms, (ii) basing decisions on the most recently received gossip results in up to a 25% reduction in selfishness, and (iii) gossiping is particularly effective at aiding agents with little or no interaction history, such as when first entering a system.

Recent work has highlighted how consideration of learning styles in the process of group formation for collaborative learning can have a positive impact. This paper investigates the performance of similar learning style groups and diverse learning style groups in group work, and explores how a grouping algorithm based on students' learning styles affects students' learning achievements and processes. An empirical study with current undergraduate students in the UK has been conducted. The participating students were invited to accomplish two group discussion tasks. Each student performed the two activities, in a separate group each time, once in a group consisting of students with similar learning styles, once in a group with diverse learning styles. No volunteer shared both groups with any individual student. This paper focuses on analysing the learning achievements and collaboration processes for the two types of groups, particularly with respect to the quality of group interactions. A significant difference was found between the percentages of time spent on meaningful interactions by the two types of groups, revealing that diverse learning style groups tend to spend significantly more time on meaningful interactions than similar learning style groups. In addition, the diverse learning style groups had demonstrated significantly less negative social-emotional reactions in showing disagreements.

In this paper we investigate data mining approaches to road type classification based on CAN (controller area network) bus data collected from vehicles on UK roads. We consider three related classification problems: road type (A, B, C and Motorway), signage (None, White, Green and Blue) and carriageway type (Single or Double). Knowledge of these classifications has a number of uses, including tuning the engine and adapting the user interface according to the situation. Furthermore, the current road type and surrounding area gives an indication of the driver's workload. In a residential area the driver is likely to be overloaded, while they may be under stimulated on a highway. Several data mining and temporal analysis techniques are investigated, along with selected ensemble classifiers and initial attempts to deal with a class imbalance present in the data. We find that the Random Forest ensemble algorithm has the best performance, with an AUC of 0.89 when used with a wavelet-Gaussian summary of the previous 2.5 seconds of speed and steering wheel angle recordings. We show that this technique is at least as good as a model-based solution that was manually created using domain expertise.

In this paper we propose a market-based approach for seeding recommendations for new items in which publishers bid to have items presented to the most influential users for each item. Users are able to select (or not) items for rating on an earn-per-rating basis, with payment given for providing a rating regardless of whether the rating is positive or negative. This approach reduces the time taken to obtain ratings for new items, while encouraging users to give honest ratings (to increase their influence) which in turn supports the quality of recommendations. To support this approach we propose techniques for inferring the social influence network from users' rating vectors and recommendation lists. Using this influence network we apply existing heuristics for estimating a user's influence, adapting them to account for the new items already presented to a user. We also propose an extension to Chen et al.'s Degree Discount heuristic [Chen et al. 2009], to enable it to be used in this context. We evaluate our approach on the MovieLens dataset and show that we are able to reduce the time taken to achieve coverage, while supporting the quality of recommendations.

A nationwide sample of university students completed a survey that asked questions about their perceptions of the factors that can cause problems which exist with online or general group work. Data were obtained from 173 students at more than 18 different universities in the United Kingdom. Three main problems that exist in group collaboration are identified through an extensive review of literature and addressed as problem scenarios in the survey. These include: poor motivation, lack of individual accountability and negative interdependence. Findings from the survey include that on average more than five factors are acknowledged by the students for each subcategory of problem, and for each scenario the factors that affect the group work are ranked by importance level. Furthermore, we find no statistically significant association between the students' backgrounds and their perceptions of the factors identified.

One of the factors that affect successful collaborative learning is the composition of collaborative groups. Due to the lack of intelligent grouping according to learnerspsila pedagogic needs in current online collaborative learning environments, developing intelligent grouping according to individual learnerspsila cognitive characteristics is highly desired. In this paper, we propose a new approach to supporting intelligent grouping based on learnerspsila learning styles. Our approach achieves the balance of different levels of learning styles in group composition. We demonstrate how it can fit into current activity-based collaborative learning environments and how it could be applied in a real world application.

Distributed systems generally require their component parts to interact cooperatively, in order for the system as a whole to function effectively. For any given activity, several alternative components may have the required capabilities. However, these components may be unreliable or dishonest, and are typically locally controlled. We view such systems as multi-agent systems, comprising autonomous agents that must cooperate for the system to be effective. In this paper we propose a mechanism, called MDT-R, for agents to delegate activities appropriately, using trust and the recommendations of their peers to meet preferences such as minimizing risk and maximizing quality.

Few intelligent learning systems exist which are dynamic and able to provide personalized learning materials to satisfy individual students' requirements. We have developed an agent-based learning system that incorporates learning objects to facilitate personalization, and is based on a learning style theory as the pedagogic foundation for adaptivity. In this paper, we present our novel approach to the incorporation of learning style theory and learning objects, and evaluation indicates that the approach is able to provide personalized learning materials and improve the adaptivity in learning systems.

Computer programming lends itself to automated assessment. With appropriate software tools, program correctness can be measured, along with an indication of quality according to a set of metrics. Furthermore, the regularity of program code allows plagiarism detection to be an integral part of the tools that support assessment. In this paper, we describe a submission and assessment system, called BOSS, that supports coursework assessment through collecting submissions, performing automatic tests for correctness and quality, checking for plagiarism, and providing an interface for marking and delivering feedback. We describe how automated assessment is incorporated into BOSS such that it supports, rather than constrains, assessment. The pedagogic and administrative issues that are affected by the assessment process are also discussed.

The open grid services infrastructure (OGSI) defines a distributed system framework by integrating grid and Web services technologies to facilitate resource sharing. In OGSI, Web services are supplemented with additional features in order to meet the requirements of grid computing. However, the issue of grid service composition is not well addressed in the OGSI framework. We apply BPEL4WS (business process execution language for Web services) as a business workflow description language for the composition of grid services. We provide an in depth analysis of BPEL4WS and OGSI in terms of their similarities and differences in areas such as life cycle management, Web service instantiation and instance group management. Based on our analysis we propose a high-level architecture to compliment OGSI with BPEL4WS for defining process workflow among grid services. We describe a prototype system which shows how the proposed architecture can be used in modelling or orchestrating grid services with BPEL4WS.

The development of an online submission system at the authors' computer science department over the last few years is described. The changing technologies used by the in-house software are discussed, together with the pedagogic and administrative issues that are affected by the process. The authors conclude by examining the issues that currently direct the future development of the software.