The development of artificial intelligence has likely reached an inflection point, with significant implications for how research needs to address emerging technologies and how they drive long-term socioeconomic development of importance for climate change scenarios.

This article explores the complex challenges of co-designing an AI- and learning analytics (LA)-integrated learning management system (LMS). While co-design has been proposed as a human-centred design approach for scaling AI and LA adoption, our understanding of how these design processes play out in real-life settings remains limited. This study is based on ethnographic fieldwork in primary and secondary schools and employs a relational materialist approach to trace, visualise, and analyse the increasingly complex and transformative relations between a growing number of actors. The findings shed light on the intricate ecosystem in which AI and LA are being introduced and on the marketisation of K-12 education. Instead of following a rational and sequential approach that can be easily executed, the co-design process emerged as a series of events, shifting from solely generating ideas with teachers to integrating and commercialising the LMS into a school market with an already high prevalence of educational technology (EdTech). AI and LA in education, co-design and data-driven schooling served as negotiating ideas, boundary objects, which maintained connectivity between actors, despite limited AI and LA implementation and the development of a stand-alone app. Even though teachers and students were actively involved in the design decisions, the co-design process did not lead to extensive adoption of the LMS nor did it sufficiently address the ethical issues related to the unrestricted collection of student data.

The concept of transparency is fragmented in artificial intelligence (AI) research, often limited to transparency of the algorithm alone. We propose that AI transparency operates on three levels-algorithmic, interaction, and social-all of which need to be considered to build trust in AI. We expand upon these levels using current research directions, and identify research gaps resulting from the conceptual fragmentation of AI transparency highlighted within the context of the three levels.

One-step reinforcement learning explanation methods account for individual actions but fail to consider the agents future behavior, which can make their interpretation ambiguous. We propose to address this limitation by providing hierarchical goals as context for one-step explanations. By considering the current hierarchical goal as a context, one-step explanations can be interpreted with higher certainty, as the agents future behavior is more predictable. We combine reward decomposition with hierarchical reinforcement learning into a novel explainable reinforcement learning framework, which yields more interpretable, goal-contextualized one-step explanations. With a qualitative analysis of one-step reward decomposition explanations, we first show that their interpretability is indeed limited in scenarios with multiple, different optimal policies-a characteristic shared by other one-step explanation methods. Then, we show that our framework retains high interpretability in such cases, as the hierarchical goal can be considered as context for the explanation. To the best of our knowledge, our work is the first to investigate hierarchical goals not as an explanation directly but as additional context for one-step reinforcement learning explanations.

Machine learning and other artificial intelligence (AI) technologies are predicted to play a transformative role in primary education, where these technologies for automation and personalization are now being introduced to classroom instruction. This article explores the rationales and practices by which machine learning and AI are emerging in schools. We report on ethnographic fieldwork in Sweden, where a machine learning teaching aid in mathematics, the AI Engine, was tried out by 22 teachers and more than 250 primary education students. By adopting an Actor-Network Theory approach, the analysis focuses on the interactions within the network of heterogeneous actors bound by the AI Engine as an obligatory passage point. The findings show how the actions and accounts emerging within the complex ecosystem of human actors compensate for the unexpected and undesirable algorithmic decisions of the AI Engine. We discuss expectations about AI in education, contradictions in how the AI Engine worked and uncertainties about how machine learning algorithms ‘learn’ and predict. These factors contribute to our understanding of the potential of automation and personalisation—a process that requires continued re-negotiations. The findings are presented in the form of a fictional play in two acts, an ethnodrama. The ethnodrama highlights controversies in the use of AI in education, such as the lack of transparency in algorithmic decision-making—and how this can play out in real-life learning contexts. The findings of this study contribute to a better understanding of AI in primary education.

The recent paper “Reward is Enough” by Silver, Singh, Precup and Sutton posits that the concept of reward maximisation is sufficient to underpin all intelligence, both natural and artificial, and provides a suitable basis for the creation of artificial general intelligence. We contest the underlying assumption of Silver et al. that such reward can be scalar-valued. In this paper we explain why scalar rewards are insufficient to account for some aspects of both biological and computational intelligence, and argue in favour of explicitly multi-objective models of reward maximisation. Furthermore, we contend that even if scalar reward functions can trigger intelligent behaviour in specific cases, this type of reward is insufficient for the development of human-aligned artificial general intelligence due to unacceptable risks of unsafe or unethical behaviour.

Real-world sequential decision-making tasks are generally complex, requiring trade-offs between multiple, often conflicting, objectives. Despite this, the majority of research in reinforcement learning and decision-theoretic planning either assumes only a single objective, or that multiple objectives can be adequately handled via a simple linear combination. Such approaches may oversimplify the underlying problem and hence produce suboptimal results. This paper serves as a guide to the application of multi-objective methods to difficult problems, and is aimed at researchers who are already familiar with single-objective reinforcement learning and planning methods who wish to adopt a multi-objective perspective on their research, as well as practitioners who encounter multi-objective decision problems in practice. It identifies the factors that may influence the nature of the desired solution, and illustrates by example how these influence the design of multi-objective decision-making systems for complex problems.

n/a

The high operational cost of aircraft, limited availability of air space, and strict safety regulations make training of fighter pilots increasingly challenging. By integrating Live, Virtual, and Constructive simulation resources, efficiency and effectiveness can be improved. In particular, if constructive simulations, which provide synthetic agents operating synthetic vehicles, were used to a higher degree, complex training scenarios could be realised at low cost, the need for support personnel could be reduced, and training availability could be improved. In this work, inspired by the recent improvements of techniques for artificial intelligence, we take a user perspective and investigate how intelligent, learning agents could help build future training systems. Through a domain analysis, a user study, and practical experiments, we identify important agent capabilities and characteristics, and then discuss design approaches and solution concepts for training systems to utilise learning agents for improved training value.

As the impact and importance of artificial intelligence (AI) grows, there is a growing trend to teach AI in primary and secondary education (K-12). To provide an international perspective, we have conducted three interviews with practitioners and policy makers from AI4K12 in the US (D. Touretzky, C. Gardner-McCune, and D. Seehorn), from Singapore (L. Liew) and from the European Commission (F. Benini).

AI Education for K-12 and in particular AI literacy gained huge interest recently due to the significantly influence in daily life, society, and economy. In this paper we discuss this topic of early AI education along four dimensions: (1) formal versus informal education, (2) cooperation of researchers in AI and education, (3) the level of education, and (4) concepts and tools.

Lattice-based motion planning is a hybrid planning method where a plan is made up of discrete actions, while simultaneously also being a physically feasible trajectory. The planning takes both discrete and continuous aspects into account, for example action pre-conditions and collision-free action-duration in the configuration space. Safe motion planning rely on well-calibrated safety-margins for collision checking. The trajectory tracking controller must further be able to reliably execute the motions within this safety margin for the execution to be safe. In this work we are concerned with introspective learning and reasoning about controller performance over time. Normal controller execution of the different actions is learned using machine learning techniques with explicit uncertainty quantification, for safe usage in safety-critical applications. By increasing the model accuracy the safety margins can be reduced while maintaining the same safety as before. Reasoning takes place to both verify that the learned models stays safe and to improve collision checking effectiveness in the motion planner using more accurate execution predictions with a smaller safety margin. The presented approach allows for explicit awareness of controller performance under normal circumstances, and detection of incorrect performance in abnormal circumstances. Evaluation is made on the nonlinear dynamics of a quadcopter in 3D using simulation.

Digital competence and programming are actively highlighted areas in education worldwide. They are becoming part of curricula all over the world, including the Swedish elementary school curriculum, Children are expected to develop computational thinking through programming activities, mainly in mathematics-which are supposed to be based on both proven experience and scientific grounds. Both are lacking in the lower grades of elementary school. This article gives unique insight into pupils learning during the first programming lessons based on a group of Swedish pupils experiences when entering school. The goal of the article is to inform education policy and practice. The large interdisciplinary, longitudinal research project studies approximately 1500 students aged 6-16 and their teachers over three years, using video documentation, questionnaires, and focus group interviews. This article reports on empirical data collected during the first year in one class with 30 pupils aged 6-7 years. The social semiotic, multimodal theoretical framework \"Design for Learning\" is used to investigate potential signs of learning in pupils multimodal representations when they, for example, use block programming in the primary and secondary transformation unit. We show that young pupils have positive attitudes to programming and high self-efficacy, and that pupils signs of learning in programming are multimodal and often visible in social interactions.

An important research problem in artificial intelligence is how to organize multiple agents, and coordinate them, so that they can work together to solve problems. Coordinating agents in a multi-agent system can significantly affect the systems performance-the agents can, in many instances, be organized so that they can solve tasks more efficiently, and consequently benefit collectively and individually. Central to this endeavor is coalition formation-the process by which heterogeneous agents organize and form disjoint groups (coalitions). Coalition formation often involves finding a coalition structure (an exhaustive set of disjoint coalitions) that maximizes the systems potential performance (e.g., social welfare) through coalition structure generation. However, coalition structure generation typically has no notion of goals. In cooperative settings, where coordination of multiple coalitions is important, this may generate suboptimal teams for achieving and accomplishing the tasks and goals at hand. With this in mind, we consider simultaneously generating coalitions of agents and assigning the coalitions to independent alternatives (e.g., tasks/goals), and present an anytime algorithm for the simultaneous coalition structure generation and assignment problem. This combinatorial optimization problem hasmany real-world applications, including forming goal-oriented teams. To evaluate the presented algorithms performance, we present five methods for synthetic problem set generation, and benchmark the algorithm against the industry-grade solver CPLEXusing randomized data sets of varying distribution and complexity. To test its anytime-performance, we compare the quality of its interim solutions against those generated by a greedy algorithm and pure random search. Finally, we also apply the algorithm to solve the problem of assigning agents to regions in a major commercial strategy game, and show that it can be used in game-playing to coordinate smaller sets of agents in real-time.

Robot safety is of growing concern given recent developments in intelligent autonomous systems. For complex agents operating in uncertain, complex and rapidly-changing environments it is difficult to guarantee safety without imposing unrealistic assumptions and restrictions. It is therefore necessary to complement traditional formal verification with monitoring of the running system after deployment. Runtime verification can be used to monitor that an agent behaves according to a formal specification. The specification can contain safety-related requirements and assumptions about the environment, environment-agent interactions and agent-agent interactions. A key problem is the uncertain and changing nature of the environment. This necessitates requirements on how probable a certain outcome is and on predictions of future states. We propose Probabilistic Signal Temporal Logic (ProbSTL) by extending Signal Temporal Logic with a sub-language to allow statements over probabilities, observations and predictions. We further introduce and prove the correctness of the incremental stream reasoning technique progression over well-formed formulas in ProbSTL. Experimental evaluations demonstrate the applicability and benefits of ProbSTL for robot safety.

n/a

Exploration is an important aspect of robotics, whether it is for mapping, rescue missions or path planning in an unknown environment. Frontier Exploration planning (FEP) and Receding Horizon Next-Best-View planning (RH-NBVP) are two different approaches with different strengths and weaknesses. FEP explores a large environment consisting of separate regions with ease, but is slow at reaching full exploration due to moving back and forth between regions. RH-NBVP shows great potential and efficiently explores individual regions, but has the disadvantage that it can get stuck in large environments not exploring all regions. In this work we present a method that combines both approaches, with FEP as a global exploration planner and RH-NBVP for local exploration. We also present techniques to estimate potential information gain faster, to cache previously estimated gains and to exploit these to efficiently estimate new queries.

n/a

n/a

n/a

Automated specification, generation and execution of high level missions involving one or more heterogeneous unmanned aircraft systems is in its infancy. Much previous effort has been focused on the development of air vehicle platforms themselves together with the avionics and sensor subsystems that implement basic navigational skills. In order to increase the degree of autonomy in such systems so they can successfully participate in more complex mission scenarios such as those considered in emergency rescue that also include ongoing interactions with human operators, new architectural components and functionalities will be required to aid not only human operators in mission planning, but also the unmanned aircraft systems themselves in the automatic generation, execution and partial verification of mission plans to achieve mission goals. This article proposes a formal framework and architecture based on the unifying concept of delegation that can be used for the automated specification, generation and execution of high-level collaborative missions involving one or more air vehicles platforms and human operators. We describe an agent-based software architecture, a temporal logic based mission specification language, a distributed temporal planner and a task specification language that when integrated provide a basis for the generation, instantiation and execution of complex collaborative missions on heterogeneous air vehicle systems. A prototype of the framework is operational in a number of autonomous unmanned aircraft systems developed in our research lab.

Autonomous systems situated in the real world often need to recognize, track, and reason about various types of physical objects. In order to allow reasoning at a symbolic level, one must create and continuously maintain a correlation between symbols denoting physical objects and sensor data being collected about them, a process called anchoring.In this paper we present a stream-based hierarchical anchoring framework. A classification hierarchy is associated with expressive conditions for hypothesizing the type and identity of an object given streams of temporally tagged sensor data. The anchoring process constructs and maintains a set of object linkage structures representing the best possible hypotheses at any time. Each hypothesis can be incrementally generalized or narrowed down as new sensor data arrives. Symbols can be associated with an object at any level of classification, permitting symbolic reasoning on different levels of abstraction. The approach is integrated in the DyKnow knowledge processing middleware and has been applied to an unmanned aerial vehicle traffic monitoring application.

Detecting and isolating multiple faults is a computationally expensive task. It typically consists of computing a set of tests and then computing the diagnoses based on the test results. This paper describes FlexDx, a reconfigurable diagnosis framework which reduces the computational burden while retaining the isolation performance by only running a subset of all tests that is sufficient to find new conflicts. Tests in FlexDx are thresholded residuals used to indicate conflicts in the monitored system. Special attention is given to the issues introduced by a reconfigurable diagnosis framework. For example, tests are added and removed dynamically, tests are partially performed on historic data, and synchronous and asynchronous processing are combined. To handle these issues FlexDx has been implemented using DyKnow, a stream-based knowledge processing middleware framework. Concrete methods for each component in the FlexDx framework are presented. The complete approach is exemplified on a dynamic system which clearly illustrates the complexity of the problem and the computational gain of the proposed approach.

Engineering autonomous agents that display rational and goal-directed behavior in dynamic physical environments requires a steady flow of information from sensors to high-level reasoning components. However, while sensors tend to generate noisy and incomplete quantitative data, reasoning often requires crisp symbolic knowledge. The gap between sensing and reasoning is quite wide, and cannot in general be bridged in a single step. Instead, this task requires a more general approach to integrating and organizing multiple forms of information and knowledge processing on different levels of abstraction in a structured and principled manner. We propose knowledge processing middleware as a systematic approach to organizing such processing. Desirable properties are presented and motivated. We argue that a declarative stream-based system is appropriate for the required functionality and present DyKnow, a concrete implemented instantiation of stream-based knowledge processing middleware with a formal semantics. Several types of knowledge processes are defined and motivated in the context of a UAV traffic monitoring application. In the implemented application, DyKnow is used to incrementally bridge the sense-reasoning gap and generate partial logical models of the environment over which metric temporal logical formulas are evaluated. Using such formulas, hypotheses are formed and validated about the type of vehicles being observed. DyKnow is also used to generate event streams representing for example changes in qualitative spatial relations, which are used to detect traffic violations expressed as declarative chronicles.

Research with autonomous unmanned aircraft systems is reaching a new degree of sophistication where targeted missions require complex types of deliberative capability integrated in a practical manner in such systems. Due to these pragmatic constraints, integration is just as important as theoretical and applied work in developing the actual deliberative functionalities. In this article, we present a temporal logic-based task planning and execution monitoring framework and its integration into a fully deployed rotor-based unmanned aircraft system developed in our laboratory. We use a very challenging emergency services application involving body identification and supply delivery as a vehicle for showing the potential use of such a framework in real-world applications. TALplanner, a temporal logic-based task planner, is used to generate mission plans. Building further on the use of TAL (Temporal Action Logic), we show how knowledge gathered from the appropriate sensors during plan execution can be used to create state structures, incrementally building a partial logical model representing the actual development of the system and its environment over time. We then show how formulas in the same logic can be used to specify the desired behavior of the system and its environment and how violations of such formulas can be detected in a timely manner in an execution monitor subsystem. The pervasive use of logic throughout the higher level deliberative layers of the system architecture provides a solid shared declarative semantics that facilitates the transfer of knowledge between different modules.

Any autonomous system embedded in a dynamic and changing environment must be able to create qualitative knowledge and object structures representing aspects of its environment on the fly from raw or preprocessed sensor data in order to reason qualitatively about the environment and to supply such state information to other nodes in the distributed network in which it is embedded. These structures must be managed and made accessible to deliberative and reactive functionalities whose successful operation is dependent on being situationally aware of the changes in both the robotic agent's embedding and internal environments. DyKnow is a knowledge processing middleware framework which provides a set of functionalities for contextually creating, storing, accessing and processing such structures. The framework is implemented and has been deployed as part of a deliberative/reactive architecture for an autonomous unmanned aerial vehicle. The architecture itself is distributed and uses real-time CORBA as a communications infrastructure. We describe the system and show how it can be used to create more abstract entity and state representations of the world which can then be used for situation awareness by an unmanned aerial vehicle in achieving mission goals. We also show that the framework is a working instantiation of many aspects of the JDL data fusion model.

Any autonomous system embedded in a dynamic and changing environment must be able to create qualitative knowledge and object structures representing aspects of its environment on the fly from raw or preprocessed sensor data in order to reason qualitatively about the environment. These structures must be managed and made accessible to deliberative and reactive functionalities which are dependent on being situationally aware of the changes in both the robotic agent's embedding and internal environment. DyKnow is a software framework which provides a set of functionalities for contextually accessing, storing, creating and processing such structures. The system is implemented and has been deployed in a deliberative/reactive architecture for an autonomous unmanned aerial vehicle. The architecture itself is distributed and uses real-time CORBA as a communications infrastructure. We describe the system and show how it can be used in execution monitoring and chronicle recognition scenarios for UAV applications.