Optimal performance of collaborative tasks requires consideration of the interactions between intelligent agents and their human counterparts. The functionality and success of these agents lie in their ability to maintain user trust; with too much or too little trust leading to over-reliance and under-utilisation, respectively. This problem highlights the need for an appropriate trust calibration methodology with an ability to vary user trust and decision making in-task. An online experiment was run to investigate whether stimulus difficulty and the implementation of agent features by a collaborative recommender system interact to influence user perception, trust and decision making. Agent features are changes to the Human-Agent interface and interaction style, and include presentation of a disclaimer message, a request for more information from the user and no additional feature. Signal detection theory is utilised to interpret decision making, with this applied to assess decision making on the task, as well as with the collaborative agent. The results demonstrate that decision change occurs more for hard stimuli, with participants choosing to change their initial decision across all features to follow the agent recommendation. Furthermore, agent features can be utilised to mediate user decision making and trust in-task, though the direction and extent of this influence is dependent on the implemented feature and difficulty of the task. The results emphasise the complexity of user trust in Human-Agent collaboration, highlighting the importance of considering task context in the wider perspective of trust calibration.

Optimal performance of collaborative tasks requires consideration of the interactions between socially intelligent agents, such as social robots, and their human counterparts. The functionality and success of these systems lie in their ability to establish and maintain user trust; with too much or too little trust leading to over-reliance and under-utilisation, respectively. This problem highlights the need for an appropriate trust calibration methodology, with the work in this paper focusing on the first step: investigating user trust as a behavioural prior. Two pilot studies (Study 1 and 2) are presented, the results of which inform the design of Study 3. Study 3 investigates whether trust can determine user decision making and task outcome during a human-agent collaborative task. Results demonstrate that trust can be behaviourally assessed in this context using an adapted version of the Trust Game. Further, an initial behavioural measure of trust can significantly predict task outcome. Finally, assistance type and task difficulty interact to impact user performance. Notably, participants were able to improve their performance on the hard task when paired with correct assistance, with this improvement comparable to performance on the easy task with no assistance. Future work will focus on investigating factors that influence user trust during human-agent collaborative tasks and providing a domain-independent model of trust calibration.

Robots are a distributive technology that promotes change. Human-Robot Interaction is an emerging field that has exciting prospects for widespread transformation. Today, embedding social intelligence into human-robot interaction design is a moonshot. Moonshots can boost science and engineering by galvanizing and incentivizing research communities and industry in the pursuit of impossible challenges to accelerate progress and scale impact. Social intelligence is all about power and influence; privacy and trust; ethical decisions and happiness. What if, robots could take the initiative and purposefully persuade you to believe something new. HRI with social intelligence has the potential to create breakthrough insights and that will energize the reimagination of how humans and robots will collaborate in business and society.

Building trust in robots through social interactions has a major impact on user experience and adoption of robot technologies. The role of trust in such interactions is associated with the persuasive influence a robot has on humans. A persuasive attempt may decrease trusting attitudes towards robots if it leads to persuasive backfiring, which refers to the creation of an attitude change in a direction opposite to the one intended by the intervention. In order to explore persuasive backfiring in the context of Human-Robot Interaction, this research study tests the interaction between emotion and logic as elements present both in the attitudes to be influenced, and in the persuasive appeal delivered by a robot. Results indicate a significant backfiring effect when emotions are used to influence attitudes that are based on logic. This observation has practical design implications for persuasive robots, especially in high-stakes fields such as Psychotherapy and Urban Search and Rescue.

Skill shortages are a drain on society. They hamper economic opportunities for individuals, slow growth for firms, and impede labor productivity in aggregate. Therefore, the ability to understand and predict skill shortages in advance is critical for policy-makers and educators to help alleviate their adverse effects. This research implements a high-performing Machine Learning approach to predict occupational skill shortages. In addition, we demonstrate methods to analyze the underlying skill demands of occupations in shortage and the most important features for predicting skill shortages. For this work, we compile a unique dataset of both Labor Demand and Labor Supply occupational data in Australia from 2012 to 2018. This includes data from 7.7 million job advertisements (ads) and 20 official labor force measures. We use these data as explanatory variables and leverage the XGBoost classifier to predict yearly skills shortage classifications for 132 standardized occupations. The models we construct achieve macro-F1 average performance scores of up to 83 per cent. Our results show that job ads data and employment statistics were the highest performing feature sets for predicting year-to-year skills shortage changes for occupations. We also find that features such as `Hours Worked', years of `Education', years of `Experience', and median `Salary' are highly important features for predicting occupational skill shortages. This research provides a robust data-driven approach for predicting and analyzing skill shortages, which can assist policy-makers, educators, and businesses to prepare for the future of work.

The policy of rational belief revision is encoded in the so-called AGM revision functions. Such functions are characterized (both axiomatically and constructively) within the well-known AGM paradigm, proposed by Alchourrón, Gärdenfors and Makinson. In this article, we show that - although not in a straightforward way - a sufficient extension of the underlying language allows for the modelling of any AGM revision function (defined at the initial language), by means of a Hamming-based rule for belief revision introduced by Dalal (defined at the extended language). The established results enrich the applicability of Dalal's proposal, leading to a conceptual and ontological reduction, as well as open new doors for the construction of any type of revision function in a practical context, given the intuitive appeal and simplicity of Dalal's construction.

The AGM paradigm for belief change, as originally introduced by Alchourron, Gärdenfors and Makinson, lacks any guidelines for the process of iterated revision. One of the most influential work addressing this problem is Darwiche and Pearl's approach (DP approach, for short), which, despite its well-documented shortcomings, remains to this date the most dominant. In this article, we make further observations on the DP approach. In particular, we prove that the DP postulates are, in a strong sense, inconsistent with Parikh's relevance-sensitive axiom (P), extending previous initial conflicts. Immediate consequences of this result are that an entire class of intuitive revision operators, which includes Dalal's operator, violates the DP postulates, as well as that the Independence postulate and Spohn's conditionalization are inconsistent with axiom (P). The whole study, essentially, indicates that two fundamental aspects of the revision process, namely, iteration and relevance, are in deep conflict, and opens the discussion for a potential reconciliation towards a comprehensive formal framework for knowledge dynamics.

Abstract Parikh’s relevance-sensitive axiom (P) for belief revision is open to two different interpretations, i.e. the weak and the strong version of (P), both of which are plausible depending on the context. Given that strong (P) has not received the attention it deserves, in this article, an extended examination of it is conducted. In particular, we point out interesting properties of the semantic characterization of the strong version of (P), as well as a vital feature of it that, potentially, results in a significant drop on the resources required for an implementation of a belief-revision system. Lastly, we shed light on the natural connection between global and local revision functions, via their corresponding semantic characterization, hence, a means for constructing global revision functions from local ones, and vice versa, is provided.

Teaching by demonstrations and teaching by assigning rewards are two popular methods of knowledge transfer in humans. However, showing the right behaviour (by demonstration) may appear more natural to a human teacher than assessing the learner’s performance and assigning a reward or punishment to it. In the context of robot learning, the preference between these two approaches has not been studied extensively. In this article, we propose a method that replaces the traditional method of reward assignment with action assignment (which is similar to providing a demonstration) in interactive reinforcement learning. The main purpose of the suggested action is to compute a reward by seeing if the suggested action was followed by the self-acting agent or not. We compared action assignment with reward assignment via a user study conducted over the web using a two-dimensional maze game. The logs of interactions showed that action assignment significantly improved users’ ability to teach the right behaviour. The survey results showed that both action and reward assignment seemed highly natural and usable, reward assignment required more mental effort, repeatedly assigning rewards and seeing the agent disobey commands caused frustration in users, and many users desired to control the agent’s behaviour directly.

Labour demand and skill shortages have historically been difficult to assess given the high costs of conducting representative surveys and the inherent delays of these indicators. This is particularly consequential for fast developing skills and occupations, such as those relating to Data Science and Analytics (DSA). This paper develops a data-driven solution to detecting skill shortages from online job advertisements (ads) data. We first propose a method to generate sets of highly similar skills based on a set of seed skills from job ads. This provides researchers with a novel method to adaptively select occupations based on granular skills data. Next, we apply this adaptive skills similarity technique to a dataset of over 6.7 million Australian job ads in order to identify occupations with the highest proportions of DSA skills. This uncovers 306,577 DSA job ads across 23 occupational classes from 2012-2019. Finally, we propose five variables for detecting skill shortages from online job ads: (1) posting frequency; (2) salary levels; (3) education requirements; (4) experience demands; and (5) job ad posting predictability. This contributes further evidence to the goal of detecting skills shortages in real-time. In conducting this analysis, we also find strong evidence of skills shortages in Australia for highly technical DSA skills and occupations. These results provide insights to Data Science researchers, educators, and policy-makers from other advanced economies about the types of skills that should be cultivated to meet growing DSA labour demands in the future.

Notwithstanding the extensive work on iterated belief revision, there is, still, no fully satisfactory solution within the classical AGM paradigm. The seminal work of Darwiche and Pearl (DP approach, for short) remains the most dominant, despite its well-documented shortcomings. In this article, we make further observations on the DP approach. Firstly, we prove that the DP postulates are, in a strong sense, inconsistent with Parikh's relevance-sensitive axiom (P), extending previous initial conflicts. Immediate consequences of this result are that an entire class of intuitive revision operators, which includes Dalal's operator, violates the DP postulates, as well as that the Independence postulate and Spohn's conditionalization are inconsistent with (P). Lastly, we show that the DP postulates allow for more revision polices than the ones that can be captured by identifying belief states with total preorders over possible worlds, a fact implying that a preference ordering (over possible worlds) is an insufficient representation for a belief state.

Deploying social robots applications in public spaces for conducting in the wild studies is a significant challenge but critical to the advancement of social robotics. Real world environments are complex, dynamic, and uncertain. Human-Robot interactions can be unstructured and unanticipated. In addition, when the robot is intended to be a shared public resource, management issues such as user access and user privacy arise, leading to design choices that can impact on users' trust and the adoption of the designed system. In this paper we propose a user registration and login system for a social robot and report on people's preferences when registering their personal details with the robot to access services. This study is the first iteration of a larger body of work investigating potential use cases for the Pepper social robot at a government managed centre for startups and innovation. We prototyped and deployed a system for user registration with the robot, which gives users control over registering and accessing services with either face recognition technology or a QR code. The QR code played a critical role in increasing the number of users adopting the technology. We discuss the need to develop social robot applications that responsibly adhere to privacy principles, are inclusive, and cater for a broad spectrum of people.

In this article, the epistemic-entrenchment and partial-meet characterizations of Parikh's relevance-sensitive axiom for belief revision, known as axiom (P), are provided. In short, axiom (P) states that, if a belief set $K$ can be divided into two disjoint compartments, and the new information $\varphi$ relates only to the first compartment, then the revision of $K$ by $\varphi$ should not affect the second compartment. Accordingly, we identify the subclass of epistemic-entrenchment and that of selection-function preorders, inducing AGM revision functions that satisfy axiom (P). Hence, together with the faithful-preorders characterization of (P) that has already been provided, Parikh's axiom is fully characterized in terms of all popular constructive models of Belief Revision. Since the notions of relevance and local change are inherent in almost all intellectual activity, the completion of the constructive view of (P) has a significant impact on many theoretical, as well as applied, domains of Artificial Intelligence.

Research in robotics and human-robot interaction is becoming more and more mature. Additionally, more affordable social robots are being released commercially. Thus, industry is currently demanding ideas for viable commercial applications to situate social robots in public spaces and enhance customers experience. However, present literature in human-robot interaction does not provide a clear set of guidelines and a methodology to (i) identify commercial applications for robotic platforms able to position the users» needs at the centre of the discussion and (ii) ensure the creation of a positive user experience. With this paper we propose to fill this gap by providing a methodology for the design of robotic applications including these desired features, suitable for integration by researchers, industry, business and government organisations. As we will show in this paper, we successfully employed this methodology for an exploratory field study involving the trial implementation of a commercially available, social humanoid robot at an airport.

Robots interacting with humans in public spaces often need to collect users' private information in order to provide the required services. Current privacy legislation in major jurisdictions requires organisations to disclose information about their data collection process and obtain user's consent prior to collecting privacy sensitive information. In this study, we consider a privacy-sensitive design of a data collection system for face identification. We deployed a face enrolment system on a humanoid robot with human-like gesturing and speech. We compared it with an equivalent system, in terms of capability and interactive process, on a screen-based interactive kiosk. In our previous contribution, we investigated the effects that embodiment has on users' privacy considerations. We found that an embodied humanoid robot is capable of collecting more private information from users in comparison to a disembodied interactive kiosk. However, this effect was statistically significant only when the two compared systems were using a transparent interface, i.e. an interface communicating to users the privacy policies for data processing and storage. Thus, in this work, we aim to further investigate the effects of transparency on users' privacy considerations and their experience with robot applications. We found that when comparing a non-transparent vs. transparent interface within the same system (i.e. on an embodied robot or on a disembodied kiosk) transparency does not lead to significant effects on users' privacy considerations. However, we found that transparency leads to a significantly better user experience for both systems. Therefore, our overall analyses suggest that both the interactive robot and the interactive kiosk are capable of enhancing the user experience by providing transparent information to users, which is required by privacy legislation. However, an interactive kiosk providing transparent information elicits significantly more privacy concerns in users as compared to the robot supplying the very same transparent information. This exploratory study provides conclusions that provide valuable insights for designing robot applications dealing with users privacy and it discusses the related legal implications, concluding with recommendations for privacy policymakers.

Active communication between robots and humans is essential for effective human-robot interaction. To accomplish this objective, Cloud Robotics (CR) was introduced to make robots enhance their capabilities. It enables robots to perform extensive computations in the cloud by sharing their outcomes. Outcomes include maps, images, processing power, data, activities, and other robot resources. But due to the colossal growth of data and traffic, CR suffers from serious latency issues. Therefore, it is unlikely to scale a large number of robots particularly in human-robot interaction scenarios, where responsiveness is paramount. Furthermore, other issues related to security such as privacy breaches and ransomware attacks can increase. To address these problems, in this paper, we have envisioned the next generation of social robotic architectures based on Fog Robotics (FR) that inherits the strengths of Fog Computing to augment the future social robotic systems. These new architectures can escalate the dexterity of robots by shoving the data closer to the robot. Additionally, they can ensure that human-robot interaction is more responsive by resolving the problems of CR. Moreover, experimental results are further discussed by considering a scenario of FR and latency as a primary factor comparing to CR models.

Social Robots have the potential to be of tremendous utility in healthcare, search and rescue, surveillance, transport, and military applications. In many of these applications, social robots need to advise and direct humans to follow important instructions. In this paper, we present the results of a Human-Robot Interaction field experiment conducted using a PR2 robot to explore key factors involved in obedience of humans to social robots. This paper focuses on studying how the human degree of obedience to a robot's instructions is related to the perceived aggression and authority of the robot's behavior. We implemented several social cues to exhibit and convey both authority and aggressiveness in the robot's behavior. In addition to this, we also analyzed the impact of other factors such as perceived anthropomorphism, safety, intelligence and responsibility of the robot's behavior on participants' compliance with the robot's instructions. The results suggest that the degree of perceived aggression in the robot's behavior by different participants did not have a significant impact on their decision to follow the robot's instruction. We have provided possible explanations for our findings and identified new research questions that will help to understand the role of robot authority in human-robot interaction, and that can help to guide the design of robots that are required to provide advice and instructions.

When robots and human users collaborate, trust is essential for user acceptance and engagement. In this paper, we investigated two factors thought to influence user trust towards a robot: preference elicitation (a combination of user involvement and explanation) and embodiment. We set our experiment in the application domain of a restaurant recommender system, assessing trust via user decision making and perceived source credibility. Previous research in this area uses simulated environments and recommender systems that present the user with the best choice from a pool of options. This experiment builds on past work in two ways: first, we strengthened the ecological validity of our experimental paradigm by incorporating perceived risk during decision making; and second, we used a system that recommends a nonoptimal choice to the user. While no effect of embodiment is found for trust, the inclusion of preference elicitation features significantly increases user trust towards the robot recommender system. These findings have implications for marketing and health promotion in relation to Human-Robot Interaction and call for further investigation into the development and maintenance of trust between robot and user.

In this article, we provide the epistemic-entrenchment characterization of the weak version of Parikh’s relevance-sensitive axiom for belief revision — known as axiom (P) — for the general case of incomplete theories. Loosely speaking, axiom (P) states that, if a belief set K can be divided into two disjoint compartments, and the new information φ relates only to the first compartment, then the second compartment should not be affected by the revision of K by φ. The above-mentioned characterization, essentially, constitutes additional constraints on epistemic-entrenchment preorders, that induce AGM revision functions, satisfying the weak version of Parikh’s axiom (P).

Nowadays, robots are gradually appearing in public spaces such as libraries, train stations, airports and shopping centres. Only a limited percentage of research literature explores robot applications in public spaces. Studying robot applications in the wild is particularly important for designing commercially viable applications able to meet a specific goal. Therefore, in this paper we conduct an experiment to test a robot application in a shopping centre, aiming to provide results relevant for today's technological capability and market. We compared the performance of a robot and a human in promoting food samples in a shopping centre, a well known commercial application, and then analysed the effects of the type of engagement used to achieve this goal. Our results show that the robot is able to engage customers similarly to a human as expected. However unexpectedly, while an actively engaging human was able to perform better than a passively engaging human, we found the opposite effect for the robot. In this paper we investigate this phenomenon, with possible explanation ready to be explored and tested in subsequent research.

This paper investigates the problem of how humans understand and control human-robot collaborative action and how to build natural interactions during human-robot collaborative action. We use a `pick and place' experiment to study collaborative activities between a human and a robot. The results show that even if human participants had a good understanding of the maximum reachability of the robot, they consistently take a surprisingly long time to help and assist the robot when a target object is out of its reach. We implemented a number of social cues in the experiment, analysed their effects in order to identify the role they could play to improve the fluency of human-robot collaboration. The experimental results showed that when the robot uses head movements, two hands or a gesture to indicate non-reachability, people react in a more natural way to assist the robot.

Action recognition from a single image is an important task for applications such as image annotation, robotic navigation, video surveillance and several others. Existing methods for recognizing actions from still images mainly rely on either bag-of-feature representations or pose estimation from articulated body-part models. However, the relationship between the action and the containing image is still substantially unexplored. Actually, the presence of given objects or specific backgrounds is likely to provide informative clues for the recognition of the action. For this reason, in this paper we propose approaching action recognition by first partitioning the entire image into superpixels, and then using their latent classes as attributes of the action. The action class is predicted based on a graphical model composed of measurements from each superpixel and a fully-connected graph of superpixel classes. The model is learned using a latent structural SVM approach, and an efficient, greedy algorithm is proposed to provide inference over the graph. Differently from most existing methods, the proposed approach does not require annotation of the actor (usually provided as a bounding box). Experimental results over the challenging Stanford 40 Action dataset have reported an impressive mean average precision of 72.3%, the highest achieved to date.

The aim of the movie is to highlight some of the key challenges facing social robots in the wild. The opening scene shows a PR2 leaving a research laboratory venturing into the real world alone in search of meaning. Each subsequent scene in the movie raises important research questions highlighting problems that need to be addressed in the field of social service robotics. When will robots wander around buildings unsupervised? How will they navigate and localize with glass walls: this research problem is exposed when a robot finds itself having to move around a real building.

Pointing is a typical means of directing a human's attention to a specific object or event. Robot pointing behaviours that direct the attention of humans are critical for human-robot interaction, communication and collaboration. In this paper, we describe an experiment undertaken to investigate human comprehension of a humanoid robot's pointing behaviour. We programmed a NAO robot to point to markers on a large screen and asked untrained human subjects to identify the target of the robots pointing gesture. We found that humans are able to identify robot pointing gestures. Human subjects achieved higher levels of comprehension when the robot pointed at objects closer to the gesturing arm and when they stood behind the robot. In addition, we found that subjects performance improved with each assessment task. These new results can be used to guide the design of effective robot pointing behaviours that enable more effective robot to human communication and improve human-robot collaborative performance.

Multiple Belief Change extends the classical AGM framework for Belief Revision introduced by Alchourron, Gardenfors, and Makinson in the early ’80s. The extended framework includes epistemic input represented as a (possibly infinite) set of sentences , as opposed to a single sentence assumed in the original framework. The transition from single to multiple epistemic input worked out well for the operation of belief revision. The AGM postulates and the system-of-spheres model were adequately generalized and so was the representation result connecting the two. In the case of belief contraction however, the transition was not as smooth. The generalized postulates for contraction, which were shown to correspond precisely to the generalized partial meet model , failed to match up to the generalized epistemic entrenchment model . The mismatch was fixed with the addition of an extra postulate, called the limit postulate , that relates contraction by multiple epistemic input to a series of contractions by single epistemic input. The new postulate however creates problems on other fronts. First, the limit postulate needs to be mapped into appropriate constraints in the partial meet model. Second, via the Levi and Harper Identities, the new postulate translates into an extra postulate for multiple revision, which in turn needs to be characterized in terms of systems of spheres. Both these open problems are addressed in this article. In addition, the limit postulate is compared with a similar condition in the literature, called (K*F), and is shown to be strictly weaker than it. An interesting aspect of our results is that they reveal a profound connection between rationality in multiple belief change and the notion of an elementary set of possible worlds (closely related to the notion of an elementary class of models from classical logic).

Intelligent agents require methods to revise their epistemic state as they acquire new information. Jeffrey's rule, which extends conditioning to probabilistic inputs, is appropriate for revising probabilistic epistemic states when new information comes in the form of a partition of events with new probabilities and has priority over prior beliefs. This paper analyses the expressive power of two possibilistic counterparts to Jeffrey's rule for modeling belief revision in intelligent agents. We show that this rule can be used to recover several existing approaches proposed in knowledge base revision, such as adjustment, natural belief revision, drastic belief revision, and the revision of an epistemic state by another epistemic state. In addition, we also show that some recent forms of revision, called improvement operators, can also be recovered in our framework.

In this paper we explore the potential of using extended belief change operators for modeling the evolution of legal evidence. We introduce a new representation, an evidence structure that can be used to support rich change operators called transmutations. Evidence structures and transmutations support the iterated nature of evidence gathering and acquisition. Importantly reliability is maintained during transmutations and modified where necessary using the Principle of Minimal Change. We establish that this process possesses desirable properties; in particular we construct transmutations that do not assume logical omniscience and that satisfy the widely accepted AGM rationality postulates for revision and contraction but at the same time preserve a key value of the evidence, its reliability.

A robot may not be truly self-aware even though it can have some characteristics of self-awareness, such as having emotional states or the ability to recognize itself in the mirror. We define self-awareness in robots to be characterized by the capacity to direct attention toward their own mental state. This paper explores robot self-awareness and the role that attention plays in the achievement self-awareness. We propose a new attention based approach to self-awareness called ASMO and conduct a comparative analysis of approaches that highlights the innovation and benefits of ASMO. We then describe how our attention based self-awareness can be designed and used to develop self-awareness in state-of-the-art humanoidal robots.

This paper investigates simple syntactic methods to revise prioritized belief bases, that are semantically meaningful in the frameworks of possibility theory and of Spohn's (1988) ordinal conditional functions. Here, revising prioritized belief bases amounts to conditioning a distribution function on interpretations. Different types of scales for priorities are discussed: finite vs. infinite, numerical vs. ordinal. Syntactic revision is envisaged as a process which transforms prioritized belief bases into a new prioritized belief base, and thus allows for the subsequent iteration.

Default logic is a prominent rigorous method of reasoning with incomplete information based on assumptions. It is a static reasoning approach, in the sense that it doesn't reason about changes and their consequences. On the other hand, its nonmonotonic behaviour appears when a change to a default theory is made. This paper studies the dynamic behaviour of default logic in the face of changes, a concept that we motivate by a reference to requirements engineering. The paper defines a contraction and a revision operator, and studies their properties. This work is part of an ongoing project whose aim is to build an integrated, domain-independent toolkit of logical methods for reasoning with changing and incomplete information. The techniques described in this paper will be implemented as part of the toolkit.