Interactive Learning Tasks Research Articles

Learning Fast and Slow: A Redux of Levels of Learning in General Autonomous Intelligent Agents

Autonomous intelligent agents, including humans, operate in a complex, dynamic environment that necessitates continuous learning. We revisit our thesis that proposes that learning in human-like agents can be categorized into two levels: Level 1 (L1) involving innate and automatic learning mechanisms, while Level 2 (L2) comprises deliberate strategies controlled by the agent. Our thesis draws from our experiences in building artificial agents with complex learning behaviors, such as interactive task learning and open-world learning.

Decomposed Inductive Procedure Learning: Learning Academic Tasks with Human-Like Data Efficiency

Human brains have many differently functioning regions which play specialized roles in learning. By contrast, methods for training artificial neural networks, such as reinforcement-learning, typically learn exclusively via a single mechanism: gradient descent. This raises the question: might human learners’ advantage in learning efficiency over deep-learning be attributed to the interplay between multiple specialized mechanisms of learning? In this work we review a series of simulated learner systems which have been built with the aim of modeling human student’s inductive learning as they practice STEM procedural tasks. By comparison to modern deep-learning based methods which train on thousands to millions of examples to acquire passing performance capabilities, these simulated learners match human performance curves---achieving passing levels of performance within about a dozen practice opportunities. We investigate this impressive learning efficiency via an ablation analysis. Beginning with end-to-end reinforcement learning (1-mechanism), we decompose learning systems incrementally to construct the 3-mechanism inductive learning characteristic of prior simulated learners such as Sierra, SimStudent and the Apprentice Learner Architecture. Our analysis shows that learning decomposition plays a significant role in achieving data-efficient learning on par with human learners---a greater role even than simple distinctions between symbolic/subsymbolic learning. Finally we highlight how this breakdown in learning mechanisms can flexibly incorporate diverse forms of natural language and interface grounded instruction, and discuss opportunities for using these flexible learning capabilities in interactive task learning systems that learn directly from a user’s natural instruction.

A KMP-based interactive learning approach for robot trajectory adaptation with obstacle avoidance

Purpose Imitation learning is a powerful tool for planning the trajectory of robotic end-effectors in Cartesian space. Present methods can adapt the trajectory to the obstacle; however, the solutions may not always satisfy users, whereas it is hard for a nonexpert user to teach the robot to avoid obstacles in time as he/she wishes through demonstrations. This paper aims to address the above problem by proposing an approach that combines human supervision with the kernelized movement primitives (KMP) model. Design/methodology/approach This approach first extracts the reference database used to train KMP from demonstrations by using Gaussian mixture model and Gaussian mixture regression. Subsequently, KMP is used to modulate the trajectory of robotic end-effectors in real time based on feedback from its interaction with humans to avoid obstacles, which benefits from a novel reference database update strategy. The user can test different obstacle avoidance trajectories in the current task until a satisfactory solution is found. Findings Experiments performed with the KUKA cobot for obstacle avoidance show that this approach can adapt the trajectories of the robotic end-effector to the user’s wishes in real time, including trajectories that the robot has already passed and has not yet passed. Simulation comparisons also show that it exhibits better performance than KMP with the original reference database update strategy. Originality/value An interactive learning approach based on KMP is proposed and verified, which not only enables users to plan the trajectory of robotic end-effectors for obstacle avoidance more conveniently and efficiently but also provides an effective idea for accomplishing interactive learning tasks under constraints.

Enhancing Human-Robot Collaboration in Industry 4.0 with AI-driven HRI

Human-robot interaction (HRI) is an important consideration in mechatronic design to ensure safe and intuitive operation of robotic systems. With advancements in artificial intelligence (AI), new opportunities have emerged to enhance HRI through learned models that can adapt to human behavior and preferences. This paper provides a comprehensive review of techniques to integrate AI into HRI for mechatronic systems. An overview is first provided of challenges and objectives in integrating intelligence into robotics for effective HRI. Modern approaches utilizing neural networks, reinforcement learning, and graph neural networks are then discussed for robotic perception, decision-making, motion control, and interaction adaptation. Additionally, hybrid approaches combining rule-based methods with learned models are highlighted. Guidelines are provided for collecting human interaction data, evaluating integrated system performance, and considering adjustability, explainability, and safety. Multiple tables summarize key studies on AI-enhanced user interfaces, interactive task learning, socially aware navigation, bio-inspired sensorimotor control, and personalized robots. Finally, open issues and future outlook are discussed. This paper aims to support mechatronic designers through an structured analysis of the emerging field of intelligent HRI with insights into current best practices for integration.

The Influence of Interactive Learning Tasks on the Motivation of Students in the Study of Chemistry

In the work were examined the possibilities of using interactive learning tasks (ILT) in the educational process when teaching chemistry: in lessons, in individual group work, when school students do homework and when preparing for exams. During the pedagogical experiment, the hypothesis was confirmed that the use of ILT for teaching chemistry helps to increase school students' interest in studying chemistry, activates their mental activity, facilitates the assimilation of the studied material and its memorization. Various forms of interactive learning tasks for teaching chemistry have been tested in the educational process. The comparison of the effectiveness of open and closed interactive learning tasks showed that the use of open ILT in the educational process contributes to improving the quality and strength of students' knowledge to a greater extent than the use of closed ILT. To find out the motivation of educational activity, the T.D. Dubovitskaya's method of diagnosing orientation of motivation was used. Experimentally confirmed the creative realization of knowledge, skills and abilities as a result of deepening the computer knowledge of school students, improving the quality of knowledge and interest through interactive learning tasks in chemistry. The use of interactive tests was aimed at the formation of educational and cognitive, value-semantic, informational and communicative competencies. The results of the survey of students confirm the expediency of using interactive tests with the formation of students' competencies. It is shown that 95% of school students like to work on interactive learning tasks, and the vast majority of students show interest in studying chemistry.

Individualization, Differentiation and Interactivity - Paradigms of Effective Learning

Centering on the student in the context of the instructional-educational action is achieved by generating both interactive/collaborative and differentiated/individualized learning situations, all the more so as the appropriate technological resources can be exploited to provide the feedback necessary for adjustment in learning. The present study presents an investigative analysis to decipher the understanding of the concepts of didactic and exploitative individualization, differentiation and interactivity, with the aim of identifying the contribution of differentiation/individualization and interactivity in achieving effective learning. The study revealed a common perception of the concepts of individualization, differentiation and interactivity at the level of teachers for preschool, primary and secondary education. Also, the academic performance of students in pre-university education (preschool, primary and secondary school) was influenced by interactive strategies and learning tasks based on individualization and differentiation to a comparative extent with their contribution to the development of the degree of socialization and positive interaction, of increasing the level social-emotional skills.

Interactively learning behavior trees from imperfect human demonstrations.

Introduction: In Interactive Task Learning (ITL), an agent learns a new task through natural interaction with a human instructor. Behavior Trees (BTs) offer a reactive, modular, and interpretable way of encoding task descriptions but have not yet been applied a lot in robotic ITL settings. Most existing approaches that learn a BT from human demonstrations require the user to specify each action step-by-step or do not allow for adapting a learned BT without the need to repeat the entire teaching process from scratch. Method: We propose a new framework to directly learn a BT from only a few human task demonstrations recorded as RGB-D video streams. We automatically extract continuous pre- and post-conditions for BT action nodes from visual features and use a Backchaining approach to build a reactive BT. In a user study on how non-experts provide and vary demonstrations, we identify three common failure cases of an BT learned from potentially imperfect initial human demonstrations. We offer a way to interactively resolve these failure cases by refining the existing BT through interaction with a user over a web-interface. Specifically, failure cases or unknown states are detected automatically during the execution of a learned BT and the initial BT is adjusted or extended according to the provided user input. Evaluation and results: We evaluate our approach on a robotic trash disposal task with 20 human participants and demonstrate that our method is capable of learning reactive BTs from only a few human demonstrations and interactively resolving possible failure cases at runtime.

‘It becomes increasingly complex to deal with multiple channels’: materialised communicative competence and digital inequality in English-medium higher education in the digital era

ABSTRACT This article explores university students’ communicative competence for English-Medium Instruction (EMI) at a Swedish university in the era of digitalisation and blended learning. Based on a linguistic ethnography, we present an argument for communicative competence as repertoire assemblages orchestrating digital materiality and human language to construct meanings. The study shows how diverse digital multimodalities and AI-language tools are essential features of spatial repertoires for academic communication, and how they cooperate with and mediate students’ personal repertoires to accomplish interactive learning tasks in EMI contexts. The study also highlights how digital diversity in EMI causes a ‘digital divide’, potentially impacting power relations among students. These findings underline the importance of acknowledging the communicative value of digital materiality and negotiating difference and normativity for intercultural academic communication in EMI.

Intercultural Language Education: Supportive Factors and Constraints on EFL Learners’ Intercultural Communicative Competence Development

This qualitative research aims to explore the supportive factors and constraints on the development of EFL learners’ intercultural communicative competence (ICC) in the EFL context of Vietnam. Forty-seven EFL learners who were learning General English at a foreign language center in Ho Chi Minh City, Vietnam, were involved in this study. Two research instruments, namely diary and semi-structured group interview, were employed to garner data. The findings revealed that there were more supportive factors (active learners, inspirational teacher, interactive learning material and task, and dynamic context) than constraints (learner differences and task administration) that affected learners’ ICC development. This study recommends that different factors such as learners, teacher, learning material/task, and context, should be scrutinized so as to facilitate the development process of learners’ ICC.

Κοινότητες Μάθησης στην Ελληνική Διασπορά: Ο Ρόλος της Τηλεδιάσκεψης

This paper explores the possibilities of using video-conferencing as part of a sister class project in order to increase the use of Greek as a second or foreign language. The research described in this paper is part of a European Union-funded project which is run by the Centre of Migration and Intercultural Studies of the University of Crete and is supervised by the Greek Ministry of Education. As a result of this program, an electronic learning environment was designed to enable the teaching and learning of Greek language and history among pupils (K-12) who attend various forms of Greek-language education worldwide. The web-based environment allows students and teachers to engage in interactive and multimodal learning tasks promoting language learning. It also gives teachers and students the opportunity to become part of a community of learning through global learning networks and to use what they have learned in creative ways and in ways that have meaning for the learners. This program has been in effect for three years, from 2011 to this date. This paper will focus on one of the twelve twinned-classrooms which took part in the Communities of Learning during the 2012-2013 school year. This case study draws conclusions dealing with the benefits of video conferencing for second/foreign language learning.

Enhancing Student Engagement and Structured Learning in Online Discussion Forums

Discussion forums remain an important component in student engagement and learning in many online courses, but current research suggests that refining their pedagogical design could significantly enhance their effectiveness. The Worksheet Video Walk Formula offers one such option. Drawing on scaffolding and question techniques, the worksheet structures the interactive learning tasks closely around the concepts and skills to be learned, while the video walk reinforces students’ connection and engagement with the task activities. This article discusses the pedagogical challenges in designing and implementing the Formula in a large (104 students) second-year English literature course on the short story. The impact of the Formula on the quality of student interaction with concepts and texts as well as their development of close reading and critical thinking skills is assessed, and future applications are explored. This study has implications for teacher presence in discussion forums, learning outcomes, student satisfaction, and online pedagogy in large classes.

Tell Me a Story! Blending Digital Storytelling Into Marketing Higher Education for Student Engagement

Multimodal learning via the use of smart devices, online social interactions, and intuitive communication platforms are fundamentally changing teaching and learning settings. Consequently, educators face unique challenges around student engagement as learners increasingly look to the use of technology-enabled activities for meaningful collaboration. Within this context, this research explores the role of digital storytelling in promoting perceived individual student and group engagement, and how these interact with group functioning, in a postgraduate marketing subject. We evaluate the implementation of this assessment as an avenue to optimize the benefits of the blended learning setting. It contributes to understanding by incorporating the concept of collective, or collaborative engagement where there is a scarcity of research despite the widespread application of group assessment in business education, and marketing education in particular. We employed canonical correlation analysis as an exploratory technique to gain initial insight into its efficacy for student engagement. We contend that digital storytelling shows promise as an inherently social and interactive learning task to provide an authentic assessment for a range of marketing problems. In turn, these attributes provide a stimulating vehicle for student engagement that can promote learning and satisfaction.

Interdisciplinary interactive blended learning concept in times of a pandemic - pain medicine "totally digital".

Introduction: Pain medicine is located in different sections of the medical curriculum. In the pandemic situation, an online teaching concept for Q14 which includes several disciplines had to be developed. The goal of the project was to create a fully digitized learning platform for the cross-sectional area Q14 that allows all participating disciplines to address the various learning goals without losing a practical component.Project description: First, the students' expectations regarding education in the field of pain medicine were recorded by means of a survey among medical students. Based on this, a teaching module in a blended learning format was developed, which consisted of two parts. Within a digital learning platform, students were first required to complete consecutive learning units using an interactive learning management system. This was followed by a presence phase (online ZOOM seminar) in which, under the guidance of teaching staff, the therapy suggestions of the individual case studies from the previous learning program were reflected. In the second part, the acquired knowledge was applied to a simulated patient. An evaluation of the online module was carried out through free-text answers and self-assessment of the completion time. The ZOOM seminar was evaluated on the basis of an assessment by the teachers. Results: The survey among students revealed a desire for practical training without “frontal teaching”. The resulting project realized this aspect by teaching theory during an online module with case vignettes and interactive learning tasks. The subsequent online presence time during the ZOOM meeting enabled the students to repeat and deepen contents and to ask questions. 170 students completed the entire online program, of which evaluation data were available for 75 students. Self-assessment of completion time averaged at 4-6 hours. In the feedback, 90 aspects were addressed, including mainly comments on content (43%), praise (33%) and comments on technical problems (23%). According to the assessment of the presenters, the students were able to carry out the pain anamnesis survey in a structured manner. The submission of the therapy proposal, however, represents a particular hurdle.Conclusion: With the presented blended learning concept it is possible to address the different learning goals and the interdisciplinarity of Q14 sufficiently. After further processing and improvement of the project, a controlled and more extensive collection of evaluation data is required to further investigate the benefit of the platform for the students regarding achievement of defined learning goals.

A Systematic Review on Oral Interactions in Robot-Assisted Language Learning

Although educational robots are known for their capability to support language learning, how actual interaction processes lead to positive learning outcomes has not been sufficiently examined. To explore the instructional design and the interaction effects of robot-assisted language learning (RALL) on learner performance, this study systematically reviewed twenty-two empirical studies published between 2010 and 2020. Through an inclusion/exclusion procedure, general research characteristics such as the context, target language, and research design were identified. Further analysis on oral interaction design, including language teaching methods, interactive learning tasks, interaction processes, interactive agents, and interaction effects showed that the communicative or storytelling approach served as the dominant methods complemented by total physical response and audiolingual methods in RALL oral interactions. The review provides insights on how educational robots can facilitate oral interactions in language classrooms, as well as how such learning tasks can be designed to effectively utilize robotic affordances to fulfill functions that used to be provided by human teachers alone. Future research directions point to a focus on meaning-based communication and intelligibility in oral production among language learners in RALL.

Characterizing an Analogical Concept Memory for Architectures Implementing the Common Model of Cognition

Architectures that implement the Common Model of Cognition - Soar, ACT-R, and Sigma - have a prominent place in research on cognitive modeling as well as on designing complex intelligent agents. In this paper, we explore how computational models of analogical processing can be brought into these architectures to enable concept acquisition from examples obtained interactively. We propose a new analogical concept memory for Soar that augments its current system of declarative long-term memories. We frame the problem of concept learning as embedded within the larger context of interactive task learning (ITL) and embodied language processing (ELP). We demonstrate that the analogical learning methods implemented in the proposed memory can quickly learn a diverse types of novel concepts that are useful not only in recognition of a concept in the environment but also in action selection. Our approach has been instantiated in an implemented cognitive system \textsc{Aileen} and evaluated on a simulated robotic domain.

Interactive Instructional: Theoretical Perspective and Its Potential Support in Stimulating Students’ Higher Order Thinking Skills (HOTS)

Interactive Instructional: Theoretical Perspective and Its Potential Support in Stimulating Students’ Higher Order Thinking Skills (HOTS)

Robots that Help Humans Build Better Mental Models of Robots

Interactive Task Learning (ITL) is an approach to teaching robots new tasks through language and demonstration. It relies on the fact that people have experience teaching each other. However, this can be challenging if the human instructor does not have an accurate mental model of a robot. This mental model consists of the robot’s knowledge, capabilities, shortcomings, goals, and intentions. The research question that I investigate is “How can the robot help the human build a better mental model of the robot?”

Expanding Task Diversity in Explanation-Based Interactive Task Learning

Expanding Task Diversity in Explanation-Based Interactive Task Learning

Interactive task learning via embodied corrective feedback

This paper addresses a task in Interactive Task Learning (Laird et al. IEEE Intell Syst 32:6–21, 2017). The agent must learn to build towers which are constrained by rules, and whenever the agent performs an action which violates a rule the teacher provides verbal corrective feedback: e.g. “No, red blocks should be on blue blocks”. The agent must learn to build rule compliant towers from these corrections and the context in which they were given. The agent is not only ignorant of the rules at the start of the learning process, but it also has a deficient domain model, which lacks the concepts in which the rules are expressed. Therefore an agent that takes advantage of the linguistic evidence must learn the denotations of neologisms and adapt its conceptualisation of the planning domain to incorporate those denotations. We show that by incorporating constraints on interpretation that are imposed by discourse coherence into the models for learning (Hobbs in On the coherence and structure of discourse, Stanford University, Stanford, 1985; Asher et al. in Logics of conversation, Cambridge University Press, Cambridge, 2003), an agent which utilizes linguistic evidence outperforms a strong baseline which does not.

ЗАСОБИ РЕАЛІЗАЦІЇ ШВИДКОГО ЗВОРОТНОГО ЗВ’ЯЗКУ В ІНФОРМАЦІЙНОМУ ОСВІТНЬОМУ СЕРЕДОВИЩІ

У статті проаналізовано методи та особливості застосування сучасних онлайн технологій, елементів інтерактивного навчання в процесі реалізації оперативного зворотного зв’язку з метою активізації навчально-пізнавальної діяльності студентів, розвитку їхніх особистісних компетентностей, формування здатності до самооцінювання результатів навчання. Наголошено, що швидкий розвиток технологій і фахово зорієнтованого програмного забезпечення зумовлюють формування інформаційного навчального середовища, у якому навчання та викладання відбуваються в умовах широкого застосування інформаційно-комунікаційних технологій. Зазначено, що вимогами ринкового середовища до конкурентоздатності фахівців стають не тільки високий рівень засвоєного студентами теоретичного матеріалу та навичок практичної роботи, але й розвиток таких особистісних якостей, як аналітичне та критичне мислення, здатність до саморозвитку, уміння ефективно спілкуватись та вирішувати конфліктні ситуації тощо. Обґрунтовано необхідність постійного осучаснення навчально-методичних комплексів дисциплін з метою організації усвідомленої та результативної навчальної діяльності студентів, формування їхньої спроможності самостійно оцінювати отримані результати. Наведено приклади використання навчальних завдань з дисципліни «Прикладна інформатика», розроблених за авторською методикою, які містять швидкий зворотний зв’язок, зорієнтовані на активізацію навчальної діяльності студентів та сприяють трансформації пасивного засвоєння ними знань в їх активне застосування у квазіпрофесійних і реальних ситуаціях навчальної діяльності. Розкрито особливості використання хмарних та онлайн технологій при здійсненні тестування рівня навчальної успішності студентів, проведення опитування за допомогою Google-форм, застосування відеоуроків у навчальній діяльності.