Computing and Information Systems - Theses
Permanent URI for this collection
Search Results
1 - 10 of 365
-
ItemNetwork Architecture for Prediction of Emergence in Complex Biological Systems( 2022)Emergence of properties at the system level, where these properties are not observed at the individual entity level, is an important feature of complex systems. Biological system emergent properties have critical roles in the functioning of organisms and the disruptions to normal functioning, and are relevant to the treatment of diseases like cancer. Complex biological systems can be modeled by abstractions in the form of molecular networks like gene regulatory networks (GRNs) and signaling networks with nodes representing molecules like genes and edges representing molecular interactions. The thesis aims at exploring the use of the architecture of these networks to predict emergence of system properties. First, to better infer the network architecture with aspects that can be useful in predicting emergence, we propose a novel algorithm Polynomial Lasso Bagging or PoLoBag for signed GRN inference from gene expression data. The GRN edge signs represent the nature of the regulatory relationships, activating or inhibitory. Our algorithm gives more accurate signed inference compared to state-of-the-art algorithms, and overcomes their weaknesses by also inferring edge directions and cycles. We also show how combining signed GRN architecture with dynamical information in our proposed dynamical K-core method predicts emergent states of the gene regulatory system effectively. Second, we investigate the existence of the bow-tie architectural organization in the GRNs of species of widely varying complexity. Prior work has shown the existence of this bow-tie feature in the GRNs of only some eukaryotes. Our investigation covers GRNs of prokaryotes to unicellular and multicellular eukaryotes. We find that the observed bow-tie architecture is a characteristic feature of GRNs. Based on differences that we observe in the bow-tie architectures across species, we predict a trend in the emergence of the dynamical gene regulatory system property of controllability with varying species complexity. Third, from input genotype data we predict an emergent phenotype at the organism level – the cancer-specific survival risk. We propose a novel Mutated Pathway Visible Neural Network or MPVNN, designed using prior knowledge of signaling network architecture and additional mutation data-based edge randomization. This randomization models how known signaling network architecture changes for a particular cancer type, which is not modeled by state-of-the-art visible neural networks. We suggest that MPVNN performs cancer-specific risk prediction better than other similar sized NN and non-NN survival analysis methods, while also providing reliable interpretations of the predictions. These three research contributions taken together make significant advances towards our goal of using molecular network architecture for better prediction of emergence, which can inform treatment decisions and lead to novel therapeutic approaches and is of value to computational biologists and clinicians.
-
ItemSoftware testing with Monte Carlo tree search( 2022)Achieving high code coverage in modern software is an essential yet challenging task. The size and complexity of programs give rise to the exploitation-exploration dilemma for testers: How to balance the consideration between investigating a) the parts of the program that appears to be interesting based on the existing information (exploitation) vs b) other parts that are less well-understood (exploration)? The Monte Carlo tree search (MCTS) is an artificial intelligence (AI) search algorithm that offers a principled solution to balancing these two concerns and has demonstrated its efficiency in large search spaces. This thesis contributes to tailoring MCTS to automate test generation for coverage-based testing. First, we propose Legion to assist coverage-based whitebox testing with our variation of MCTS. Legion’s APPFuzzing generalises concoilc testing and fuzzing to cover more program states with amortised computation cost: It selectively solves the path constraints of intermediate program states for seed inputs and mutates them to cover the descendants of the selected states. Legion’s variation of MCTS uses the best-first search strategy to learn the most productive program states to apply APPFuzzing, thus mitigating path explosion in symbolic execution. We compare Legion’s performance with KLEE, a famous coverage-based concoilc testing tool developed, maintained, and used by a large community across academia and industry. Legion outperformed KLEE in 7 out of 11 benchmark suites in a worldwide competition and achieved 90% of the best score in 5 suites. Then we extend Legion’s algorithm to seed selection in greybox fuzzing. This algorithm searches for stateful protocol servers’ most progressive seed input sequence. We analyse the impact of multiple server models and selection algorithms with repeated experiments. This study shows that although our algorithm has a limited impact on the overall coverage performance, it vastly increases the probability of covering some representative code blocks in servers. It highlights the bottleneck of overall coverage performance in input generation quality and fuzzing throughput. Finally, we design a contextual MCTS algorithm to reify and transfer the knowledge of past program state evaluations. Most software testing algorithms estimate a state’s productivity only with the state-specific statistics and neglect the knowledge from other states. To use the statistics more efficiently, we identify two contextual features of each program state and quantified their effect state’s productivity as feature coefficients. We then train a regression model to dynamically update the value of the coefficients to evaluate other program states based on the states’ feature values. By analysing our experiment results and reviewing others’ work, we point out the current weaknesses and corresponding future works regarding this topic.
-
ItemTraffic Optimization in The Era of Connected Autonomous Vehicles( 2021)Traffic optimization, especially at large scale, is an important and challenging problem due to the highly dynamic and unpredictable nature of traffic. The incoming era of connected autonomous vehicles (CAVs) offers a great opportunity for traffic optimization. The routes, which are followed by CAVs, can be assigned by a centralized system with better predictability. By collecting highly detailed real-time traffic data from sensors and CAVs, a traffic management system will have the full view of the entire road network, allowing it to plan traffic in a virtual world that replicates the real road network very closely. This thesis presents the problem of route assignment for CAVs, with the aim of optimizing traffic at network-level. We formulate the research problem as a streaming route assignment problem and address the intersecting routes which is a key reason causing delays and long queues at junctions. We focus on finding road network routes with few intersections for vehicles. We propose two route assignment algorithms to solve the problem, Local Detour Algorithm (LDA) and Multiple Intersection Reduction Algorithm (MIRA). LDA uses traffic information at road links or road junctions during route allocation, while MIRA enlarges the scope of traffic information to have a big picture of traffic conditions. LDA is an efficient algorithm due to searching detours in a small area around congested junctions. On the other hand, MIRA is more complicated as it searches for longer detours from not only congested junctions but also their surrounding junctions that will be congested in the near future. The results show that the route allocator can decrease traffic congestion significantly by utilizing MIRA. By assigning new routes on roads, traffic conditions can change significantly after a certain time, which means that the real-time traffic conditions will not be an accurate estimate of the near future. What if the route allocator can predict the effect of route assignments on roads? Traffic congestion prediction enables the route allocator to realize traffic conditions accurately. For a specific type of traffic road network in which intersections are signalized, we propose a predictive congestion model by modifying a real-time queue-based congestion model. The model predicts traffic congestion for each road link based on the corresponding number of route allocations. We also propose a route assignment algorithm, Traffic Congestion Aware Route Assignment Algorithm (TCARA), which uses the predictive congestion model. TCARA is designed for a specific type of road networks, while MIRA is a general route assignment algorithm with no prior assumption on road intersections (i.e., whether they are signalized or not), making it a good candidate for a wider range of road networks. Another reason that can change traffic conditions is varying traffic demand over time. Thus, if a route is optimized based on the current traffic conditions while the traffic demand is not stable, the route may be ineffective. What happens if prior temporal traffic data for particular traffic conditions of a road network is available? Temporal traffic data demonstrates historical traffic conditions collected at regular time intervals. The route allocator can be aware of changes in traffic demand that may occur in the near future. So, it can optimize traffic furthermore by having access to such data. We propose a new route assignment algorithm, Temporal Multiple Intersection Reduction Algorithm (T-MIRA), by extending MIRA to leverage prior temporal traffic data for reducing traffic congestions. In this thesis, MIRA is the main contribution. We extend it for situations when traffic conditions evolve due to the impact of route assignments or change of traffic demand. We investigate two different approaches, using a predictive congestion model in TCARA and using temporal traffic data in T-MIRA. An advanced traffic management system may utilize different approaches and solutions in an integrated way to achieve the best outcome in the real world.
-
ItemTowards dialogue system training with little labeling effort( 2022)Dialogue systems conduct natural and meaningful conversations with humans and are applied in diverse tasks including information seeking, task completion, and social chat. They have made great progress in recent years and have been widely integrated into commercial systems used by people everyday, such as virtual assistants in smartphones like Apple Siri and Google Assistant. There are also efforts to develop domain-specific dialogue systems, such as for e-commerce or education domain. However, building these dialogue systems requires great human efforts including task schema design, sample dialogue collection, and human annotation, and such a complex process may be needed for each new application. In this thesis, we aim to alleviate the costs of building a dialogue system via better exploiting both labeled and unlabeled dialogue data. We study three problems. The first is semi-supervised policy learning that aims to reduce annotation efforts for policy learning. Dialogue policy learning heavily relies on reward signals, which promote efficient task completion and penalize inappropriate decisions. Ground-truth decision making examples (i.e., expert demonstrations) are widely utilized to provide fine-grained rewards by measuring the similarity between behaviours of policies being learned and those shown in expert demonstrations. Existing approaches require labeling expert demonstrations into the form of state-action pairs, which requires extensive annotations. To reduce annotation efforts, we study semi-supervised policy learning and unify both labeled and unlabeled dialogues to provide rewards for policy learning. We propose a novel sequence model based reward estimator for policy learning which models dialogue progress (i.e., state transitions) either with or without annotations. Extensive experiments show that the proposed approach outperforms state-of-the-art approaches by up to 35.4 and 18 percents in terms of success rate and required turns, respectively, while it reduces the amount of labeled dialogue data by 80 percents. The second problem is cross-domain dialogue generation that aims to reduce dialogue collection efforts for response generation. Response generation relies on large-scale dialogue data which might be unavailable for certain domains. Although domain adaptation may alleviate the problem of lacking training data, a straightforward adaptation is easily prone to domain gap due to complex characteristics of the response generation task. To address such limitations, we break response generation into two core steps, content planning and surface realization, which are responsible for predicting response intention and ensuring language fluency, respectively. To better disentangle the two steps, we propose to learn semantic text representations that encode dialogue responses' underlying intentions, which facilitate knowledge transferring. The proposed approach achieves up to a 10.4 percents gain in task completion (measured by entity F1 score) compared to previous approaches. The third problem is data augmentation for information-seeking question answering (QA), which a special task of a dialogue system. To answer information seeking queries, the system needs to perform reasoning over external knowledge sources, e.g., textual paragraphs. Queries that require multiple reasoning steps to derive the answers, i.e., multi-hop questions, pose more challenges than those that only require a single reasoning step. However, collecting training data for multi-hop questions is costly due to complex language expressions and underlying reasoning logic. To reduce data collection efforts, we propose to generate synthetic multi-hop QA pairs using large-scale easily obtained single-hop QA data. To unify single-hop and multi-hop QA data, we propose a generate-and-compose approach, which generates sub-questions and composes them into a target multi-hop question. Experiments show that the proposed approach outperforms competitive baselines on a benchmark multi-hop QA dataset and achieves up to a 9.3 percents performance gain in terms of F1 score for a downstream QA model.
-
ItemEnergy and Time Aware Scheduling of Applications in Edge and Fog Computing Environments( 2022)The Internet of Things (IoT) paradigm is playing a principal role in the advancement of many application scenarios such as healthcare, smart city, transportation, entertainment, and agriculture, which significantly affect the daily life of humans. The smooth execution of these applications requires sufficient computing and storing resources to support the massive amount of data generated by IoT devices. However, IoT devices are resource-limited intrinsically and are not capable of efficient processing and storage of large volumes of data. Hence, IoT devices require surrogate available resources for the smooth execution of their heterogeneous applications, which can be either computation-intensive or latency-sensitive. Cloud datacenters are among the potential resource providers for IoT devices. However, as they reside at a multi-hop distance from IoT devices, they cannot efficiently execute IoT applications, especially latency-sensitive ones. Fog computing paradigm, which extends Cloud services to the edge of the network within the proximity of IoT devices, offers low latency execution of IoT applications. Hence, it can improve the response time of IoT applications, service startup time, and network congestion. Also, it can reduce the energy consumption of IoT devices by minimizing their active time. However, Fog servers are resource-limited compared to Cloud servers, preventing them from the execution of all types of IoT applications, especially extremely computation-intensive applications. Hence, Cloud servers are used to support Fog servers to create a robust computing environment with heterogeneous types of resources. Consequently, the Fog computing paradigm is highly dynamic, distributed, and heterogeneous. Thus, without efficient scheduling techniques for the management of IoT applications, it is difficult to harness the full potential of this computing paradigm for different IoT-driven application scenarios. This thesis focuses on different scheduling techniques for the management of IoT applications in Fog computing environments while considering: a) IoT devices' characteristics, b) the structure of IoT applications, c) the context of resource providers, d) the networking characteristics of the Fog servers, e) the execution cost of running IoT applications, and f) the dynamics of computing environment. This thesis advances the state-of-the-art by making the following contributions: 1. A comprehensive taxonomy and literature review on the scheduling of IoT applications from different perspectives, namely application structure, environmental architecture, optimization properties, decision engine characteristics, and performance evaluation, in Fog computing environments. 2. A distributed Fog-driven scheduling technique for network resource allocation in dense and ultra-dense Fog computing environments to optimize throughput and satisfy users' heterogeneous demands. 3. A distributed scheduling technique for the batch placement of concurrent IoT applications to optimize the execution time of IoT applications and energy consumption of IoT devices. 4. A distributed application placement and migration management technique to optimize the execution time of IoT applications, the energy consumption of IoT devices, and the migration downtime in hierarchical Fog computing environments. 5. A Distributed Deep Reinforcement Learning (DDRL) technique for scheduling IoT applications in highly dynamic Fog computing environments to optimize the execution time of IoT applications and energy consumption of IoT devices. 6. A system software for scheduling IoT applications in multi-Cloud Fog computing environments. 7. A detailed study outlining challenges and new research directions for the scheduling of IoT applications in Fog computing environments.
-
ItemAvoiding Bad Traffic: Analyzing the Past, Adapting the Present, and Estimating the Future( 2021)Mobility is essential for modern societies. However, due to the increasing demand for mobility, traffic congestion poses a significant challenge to economic growth and advancement for many cities worldwide. At the same time, the widespread availability of location-aware devices has led to a sharp increase in the amount of traffic data generated, thereby, providing an opportunity for intelligent transportation systems to emerge as one of the main cost-effective methods for traffic congestion mitigation. This boost in traffic data has led to a new generation of live navigation services that depend on traffic estimation to provide up-to-date navigation advice. Intelligent transportation systems increase the utilization of existing infrastructure and support drivers to make better navigation decisions by providing actionable traffic information. However, a fundamental shortcoming of existing intelligent navigation systems is that they do not consider the evolution of traffic and route drivers based on snapshot traffic conditions. This is especially critical in the presence of traffic incidents, where the impact of the incident introduces significant variation in the traffic conditions around the incident as things unfold. This thesis proposes three contributions focusing on traffic estimation and forecasting to help drivers avoid bad traffic, especially around traffic incidents. The first contribution is an automated traffic management service to help drivers avoid traffic events based on analyses of historical trajectory data from other drivers. Users subscribe to the service and, when a traffic event occurs, the service provides advice based on all drivers' actions during a similar traffic event in the past. We present a solution that combines a graph search with a trajectory search to find the fastest path that was taken to avoid a similar event in the past. The intuition behind our solution is that we can avoid a traffic event by following the traces of the best driver from a similar situation in the past. The second contribution is a system that uses real-time traffic information and fast traffic simulations to adapt to traffic incident impact and generate navigation advice. In this work, we use faster than real-time simulations to model the evolution of traffic events and help drivers proactively avoid congestion caused by events. The system can subscribe to real-time traffic information and estimate the traffic conditions using fast simulations without the need for historical data. We evaluate our approach through extensive experiments to test the performance and accuracy, and quality of the navigation advice of our system with real data obtained from TomTom Traffic API. For the third contribution, we propose effective deep learning models for large-scale citywide traffic forecasting. In addressing this problem, our goal is to predict traffic conditions for thousands of sites across the city. Such large-scale predictions can be used by navigation systems to help drivers avoid congestion. We propose a traffic forecasting model based on deep convolutional networks to improve the accuracy of citywide traffic forecasting. Our proposed model uses a hierarchical architecture that captures traffic dynamics at multiple spatial resolutions. We apply a multi-task learning scheme based on this architecture, which trains the model to predict traffic at different resolutions. Our model helps provide a coherent understanding of traffic dynamics by capturing short and long spatial dependencies between different regions in a city. Experimental results on real datasets show that our model can achieve competitive results while being more computationally efficient.
-
ItemComputational models of visual search and attention in natural images: from neural circuits to search strategies( 2021)Humans use visual search constantly in their daily activities, from searching for their keys to looking out for pedestrians while driving. Cortical visual attention and human eye movements are among the mechanisms that help with an effective visual search. In this thesis, we propose a theoretical model of eye movements in a visual search task as well as a computational model of neural mechanisms underlying visual attention. Towards computationally modeling human eye movements, we propose a model of target detectability in natural scenes which can be used by a Bayesian ideal searcher. The model uses a convolutional neural network as a feature extractor for extracting the features of target and background and uses signal detection theory to estimate detectability as the discriminability of the two feature distributions. We collect the detectability of a known target on 18 different backgrounds from human observers in a two-alternative forced-choice task. We use the collected detectabilities for verification and scaling of the ones estimated by our model. We further fed the collected detectabilities to a Bayesian ideal observer to predict human eye movements in a visual search task in natural scenes. We collected human eye movements on the same 18 natural backgrounds in a search for the known target and used it as the ground truth. Our model closely followed some statistical parameters of the collected eye movements, supporting the idea that humans search near-optimal in natural images. We further generalize the detectability model to any target and background and apply the Bayesian ideal observer model on a real-world dataset of pedestrians in various contexts. To the best of our knowledge, we are the first to provide an unsupervised gaze prediction algorithm on natural scenes which uses the Bayesian ideal observer and doesn't need human eye movements for training. The presented model can have potential applications in autonomous drivers. Towards computationally modeling the neural mechanisms of visual attention, we propose a large-scale computational model of attentional enhancement of visual processing, based on the idea of neural oscillations being fed back to attended features or locations in a scene. Our proposed model supports the idea of neural oscillation for mediating spatial attention and applies the same circuit for feature-based attention as well. The presented models can be used in various human gaze assisted Artificial Intelligent systems as well as enhancing our knowledge of the human visual system.
-
ItemScalable contrast pattern mining for network traffic analysis( 2021)Contrast pattern mining is a data mining technique that characterises significant changes between datasets. Contrast patterns identify nontrivial differences between the classes of a dataset, interesting changes between multiple datasets, or emerging trends in a data stream. In this thesis, we focus on the pattern of characterizing changes in Internet traffic using contrast pattern mining. For example, network managers require a compact yet informative report of significant changes in network traffic for security and performance management. In this context, contrast pattern mining is a promising approach to provide a concise and meaningful summary of significant changes in the network. However, the volume and high dimensionality of network traffic records introduce a range of challenges for contrast pattern mining. In particular, these challenges include the combinatorial search space for contrast patterns, the need to mine contrast patterns over data streams, and identifying new changes as opposed to rare recurring changes. In this thesis, we introduce a range of novel contrast mining algorithms to address these challenges. We first introduce the use of contrast pattern mining in network traffic analysis. We show that contrast patterns have strong discriminative power that make them suitable for data summarization, and finding meaningful and important changes between different traffic datasets. We also propose several evaluation metrics that reflect the interpretability of patterns for security managers. In addition, we demonstrate on real-life datasets that the vast majority of extracted patterns are pure, i.e., most change patterns correspond to either attack traffic or normal traffic, but not a mixture of both. We propose a method to efficiently extract contrast patterns between two static datasets.We extract a high-quality set of contrast patterns by using only the most specific patterns to generate a compact and informative report of significant changes for network managers. By elimination of minimal patterns in our approach, we considerably reduce the overlap between generated patterns, and by reducing the redundant patterns, we substantially improve the scalability of contrast pattern mining and achieve a significant speed-up. We also propose a novel approach to discriminate between new events and rare recurring events. Some changes in network traffic occur on a regular basis and show periodic behaviour, which are already known to network analysts. Thus, network managers may want to filter out these known recurring changes, and prioritize their focus on new changes, given their limited time and resources. Our approach to this problem is based on second order contrast pattern mining. Our work demonstrates the importance of higher order contrast pattern mining in practice, and provides an effective method for finding such higher order patterns in large datasets. Finally, based on the approaches that we introduced for contrast pattern mining in static datasets, we then propose two novel methods to extract contrast patterns over high dimensional data streams. We consider two incremental scenarios: (i) when one dataset is changing over time and the other dataset is static as a reference dataset, and (ii) when both datasets are changing over a data stream. In our approach, instead of regenerating all patterns from scratch, we reuse the previously generated contrast patterns wherever possible to mine the new set of patterns. Using this technique, we avoid expensive incremental computation and increase the efficiency and scalability of mining on dense and high dimensional data streams. As a result of this scalability, our method also can find the solutions for datasets where the other algorithms cannot. In addition to the substantial improvements in performance and scalability of our algorithm, we demonstrate that the quality of the generated patterns of our algorithm is quite comparable with the other algorithms. In summary, we propose several efficient contrast pattern mining approaches to extract significant changes between two static datasets or over a data stream. We also introduce a novel approach to identify new changes from the recurring changes. Our experimental results on different real-life datasets demonstrate the improvements in performance of our proposed algorithms compared to the existing approaches.
-
ItemThe Role of Explanations in Enhancing Algorithmic Fairness Perceptions( 2021)Decision-makers employ machine learning algorithms for decision-making to gain insights from data and make better decisions. More specifically, advanced algorithms can help organizations classify their customers and predict their behavior at the highest accuracy levels. However, the opaque nature of those algorithms has raised concerns around some potential unintended consequences they might cause, unfair decisions at the center. Unfair decisions negatively impact on both organizations and customers. Customers may lose their trust in organizations as being treated unfairly and consequently, organizations’ reputations might be put at risk. Transparency provision has been introduced to organizations as a way of addressing the issue of algorithmic opacity. One approach for transparency provision is explaining algorithms’ performance and how they reach a decision to decision-makers. Therefore, explanations can consequently influence the fairness perceptions of the decision-makers about algorithms’ decisions. Understanding how explanations and the way of discoursing them impact on the fairness perceptions of the organizational decision-makers is important. However, little research has focused on the role of explanations in enhancing fairness perceptions. I seek to address this research gap answering the question of: “How does explanation influence decision-makers’ perceptions of fairness connected with an algorithm’s decisions?” I conduct three studies to answer this question. In study 1, I conduct a conceptual study to explore the dimensions of explanations that need to be studied in understanding the impact of explanations on fairness perceptions. In study 2, I develop a research model hypothesizing the role of perspective-taking in discoursing two different explanations with decision-makers in their fairness perceptions. I conducted a 2*2 experiment to test the hypotheses. In study 3, I develop a research model hypothesizing the influence of explanations restrictiveness in the decision’s fairness perceived by the decision-makers. I conducted a 2*2 experiment to test the hypotheses. The findings of this research result in three important insights about explanations and their role in enhancing algorithmic fairness perceptions; first, I propose four dimensions of explanations that need to be considered in understanding fairness perceptions including the content types of explanations, the explanations reasoning logic, the scope of explanations and explanations discourse. Second, taking different perspectives of organization or customer in communicating different types of explanations lead to different impact on the perception of fairness about algorithm’s performance and its made decision. Third, Framing explanations in a less restrictive way creates the space for the decision-makers to be cognitively more engaged with the algorithmic decision-making and practice their own judgment about that which consequently influences on their fairness perceptions.
-
ItemAugmented Reality Annotations to Enhance Skill-Movement Acquisition in the Classroom( 2021)In a typical classroom, annotations are drawn in whiteboards to facilitate visual communication between teacher and students. It improves classroom interaction and results in enhanced learning. Annotations can add extra helpful information, guide student’s attention, and assess students’ understanding. However, annotations are not usually used in settings where students learn skilled-movements (e.g., dance, weightlifting, martial arts) because the content is the student’s movement itself and the dynamic nature of the movements . It is not a static representation of the movement on the whiteboard. Augmented Reality (AR) technology can change the skill-movement classroom, enabling annotations to leave the whiteboard and appear on/around the student performing a skilled-movement. These AR-annotations have the potential to improve teacher-student interaction in these settings, resulting in enhanced learning of skilled-movements. This thesis presents four studies investigating AR-Aannotations using the teacher-student communication model in a skilled-movement classroom that requires 3-steps: Demonstration, Performance , and Feedback. The context for these studies is physiotherapy and weightlifting. Study 1 gathered the requirements for an AR system that could enhance the practical physiotherapy classroom through a focus group and a field study. Study 2 builtds a system with the identified requirements and evaluated it in two practical physiotherapy classes. The study demonstrated that surface AR-annotations allowed teachers to guide the students’ focus of attention and assess students’ knowledge and understanding of the movement. However, the use of surface AR-annotations required the annotated object to be static, limiting AR-annotations used when movement is required. Study 3 investigated the Performance-step, where the aim is to find different types of AR-annotations to enhance the assessment of student’s performance. The study has shown that mid-air annotations (static, temporal, assisted, and morphing) can assist the teacher while assessing students’ performance. These AR-annotations can also be used on dynamic objects without requiring them to become static. Finally, Study 4 investigated the Feedback-step, where the problem is the different viewpoints of users. While annotating the student’s body, the physical worldview of the teacher and the student cannot be the same. The study a confirmationloop that enables a simple teacher-student interaction using mid-air annotations from different viewpoints. The thesis makes three contributions. The first contribution is a new skilled-movement acquisition model that adds a Knowledge-loop in which the teacher assesses the student’s movement knowledge. The teacher asks questions about the skilled-movement, and the student uses AR-annotations to demonstrate their understanding of the movement visually. The Knowledge-loop removes the need for performance to assess students’ knowledge of the skilled-movement, enhancing motivation and self-assessment. The second contribution is the addition of a morphing transition state to the current temporality dimension aspect of the taxonomy of AR-annotations since valuable information embedded in the creation process of annotations has not been considered in the field of AR-annotations. This information includes both the transition from the beginning to the end when creating an annotation as well as annotations that are morphing in nature. This additional information influences the intended message of annotations impacting the way they are perceived. The third contribution is the expansion of the understanding of egomotion and exocentric (viewpoint reference frame dimension) to switch the focus from a selfcontrolled to a shared controlled virtual viewpoint. This thesis explored the use of AR-annotations with multiple users and the challenges involved, where controlling the position of the virtual viewpoint becomes a shared task for all the users.