Learning Grasp Affordances from Vision

Monimutkaisissa ja muuttuvissa ympäristöissä työskentelevät robotit tarvitsevat kykyä manipuloida ja tarttua esineisiin. Tämä työ tutkii robottitarttumisen ja robottitartuntapis-teiden koneoppimisen aiempaa tutkimusta ja nykytilaa. Nykyaikaiset menetelmät käydään läpi, ja Le:n koneoppimiseen pohjautuva luokitin toteutetaan, koska se tarjoaa parhaan onnistumisprosentin tutkituista menetelmistä ja on muokattavissa sopivaksi käytettävissä olevalle robotille. Toteutettu menetelmä käyttää intensititeettikuvaan ja syvyyskuvaan po-hjautuvia ominaisuuksi luokitellakseen potentiaaliset tartuntapisteet. Tämän toteutuksen tulokset esitellään.

The nature of communication structures of university students in an international web-based computer-supported collaborative learning (CSCL) course

Communication, the flow of ideas and information between individuals in a social context, is the heart of educational experience. Constructivism and constructivist theories form the foundation for the collaborative learning processes of creating and sharing meaning in online educational contexts. The Learning and Collaboration in Technology-enhanced Contexts (LeCoTec) course comprised of 66 participants drawn from four European universities (Oulu, Turku, Ghent and Ramon Llull). These participants were split into 15 groups with the express aim of learning about computer-supported collaborative learning (CSCL). The Community of Inquiry model (social, cognitive and teaching presences) provided the content and tools for learning and researching the collaborative interactions in this environment. The sampled comments from the collaborative phase were collected and analyzed at chain-level and group-level, with the aim of identifying the various message types that sustained high learning outcomes. Furthermore, the Social Network Analysis helped to view the density of whole group interactions, as well as the popular and active members within the highly collaborating groups. It was observed that long chains occur in groups having high quality outcomes. These chains were also characterized by Social, Interactivity, Administrative and Content comment-types. In addition, high outcomes were realized from the high interactive cases and high-density groups. In low interactive groups, commenting patterned around the one or two central group members. In conclusion, future online environments should support high-order learning and develop greater metacognition and self-regulation. Moreover, such an environment, with a wide variety of problem solving tools, would enhance interactivity.

A multimethod examination of contributors to successful on-the-job learning for vocational students

Rapid changes in working life and competence requirements of different professions have increased interest in workplace learning. It is considered an effective way to learn and update professional skills by performing daily tasks in an authentic environment. Especially, ensuring a supply of skilled future workers is a crucial issue for firms facing tight competition and a shortage of competent employees due to the retirement of current professionals. In order to develop and make the most of workplace learning, it is important to focus on workplace learning environments and the individual characteristics of those participating in workplace learning. The literature has suggested various factors that influence adults' and professionals’ workplace learning of profession-related skills, but lacks empirical studies on contextual and individual-related factors that positively affect students' workplace learning. Workers with vocational education form a large group in modern firms. Therefore, elements of vocational students’ successful workplace learning during their studies, before starting their career paths, need to be examined. To fill this gap in the literature, this dissertation examines contributors to vocational students’ workplace learning in Finland, where students’ workplace learning is included in the vocational education and training system. The study is divided into two parts: the introduction, comprised of the overview of the relevant literature and the conclusion of the entire study, and five separate articles. Three of the articles utilize quantitative methods and two use qualitative methods to examine factors that contribute to vocational students’ workplace learning. The results show that, from the students’ perspective, attitudinal, motivational, and organizationrelated factors enhance the student’s development of professionalism during the on-the-job learning period. Specifically, the organization-related factors such as innovative climate, guidance, and interactions with seniors have a strong positive impact on the students’ perceived development of professional skills because, for example, the seniors’ guidance and provision of new viewpoints for the tasks helps the vocational students to gain autonomy at work performance. A multilevel analysis shows that of those factors enhancing workplace learning from the student perspective, innovative climate, knowledge transfer accuracy, and the students’ performance orientation were significantly related to the workplace instructors’ assessment regarding the students’ professional performance. Furthermore, support from senior colleagues and the students’ self-efficacy were both significantly associated with the formal grades measuring how well the students managed to learn necessary professional skills. In addition, the results suggest that the students’ on-the-job learning can be divided into three main phases, of which two require efforts from both the student and the on-the-job learning organization. The first phase includes the student’s application of basic professional skills, demonstration of potential in performing daily tasks, and orientation provided by the organization at the beginning of the on-the-job learning period. In the second phase, the student actively develops profession-related skills by performing daily tasks, thus learning a fluent working style while observing the seniors’ performance. The organization offers relevant tasks and follows the student’s development. The third level indicates a student who has reached the professional level described as a full occupation. The results suggest that constructing the vocational students’ successful on-the-job learning period requires feedback from seniors, opportunities to learn to manage entire work processes, self-efficacy on the part of the students, proactive behavior, and initiative in learning. The study contributes to research on workplace learning in three ways: firstly, it identifies the key individual- and organization-based factors that influence the vocational students’ successful on-the-job learning from their perspective and examines mutual relationships between these factors. Second, the study provides knowledge of how the factors related to the students’ view of successful workplace learning are associated with the workplace instructors’ perspective and the formal grades. Third, the present study finds elements needed to construct a successful on-the-job learning for the students.

Online problem-based learning : perceptions of nursing educators /

A qualitative study was conducted to detennine 5 nursing educators' perceptions about the online application of a problem-based learning strategy in undergraduate nursing education. The question asked in the study was: Can the essential elements of face-to-face problem-based learning be supported in an online format? The data for this study came from 2 individual tape-recorded interviews with each of the 5 participants over a 3-month period and from a researchjournaI. The educators felt that student-centered learning and critical thinking could be supported within an online format. However, they noted that challenges could exist in terms of developing tutor roles, fostering student self-direction, facilitating group process and connections, and incorporating a nursing philosophy of online learning. The importance of tailoring an online problem-based learning course to reflect educators' philosophies and values in nursing emerged as an important theme from the interview responses. Overall, the participants suggested that an ideal environment would blend both face-to-face and online elements and that fewer elements would be offered in the first 2 years of the nursing program. They described a hybrid model of problem-based learning in which the online component could be used to support face-to-face sessions.

Perceptions of some teachers about the similarities and/or differences in teaching online versus in a classroom environment /

The introduction of computer and communications technology, and particularly the internet, into education has opened up some new possibilities for teaching and learning. Courses designed and delivered in an online environment offer the possibility of highly interactive and individually focussed teaching and learning experiences. However, online courses also present new challenges for both teachers and students. A qualitative study was conducted to explore teachers' perceptions about the similarities and differences in teaching in the online and face-to-face (F2F) environments. Focus group discussions were held with 5 teachers; 2 teachers were interviewed in depth. The participants, 3 female and 2 male, were full-time teachers from a large College of Applied Arts & Technology in southern Ontario. Each of them had over 10 years of F2F teaching experience and each had been involved in the development and teaching of at least one online course. i - -; The study focussed on how teaching in the online environment compares with teaching in the F2F environment, what roles teachers and students adopt in each setting, what learning communities mean online and F2F and how they are developed, and how institutional policies, procedures, and infrastructure affect teaching and learning F2F and online. This study was emic in nature, that is the teachers' words determine the themes identified throughout the study. The factors identified as affecting teaching in an online environment included teacher issues such as course design, motivation to teach online, teaching style, role, characteristics or skills, and strategies. Student issues as perceived by the teachers included learning styles, role, and characteristics or skills. As well, technology issues such as a reliable infrastructure, clear role and responsibilities for maintaining the infrastructure, support, and multimedia capability affected teaching online. Finally, administrative policies and procedures, including teacher selection and training, registration and scheduling procedures, intellectual property and workload policies, and the development and communication of a comprehensive strategic plan were found to impact on teaching online. The teachers shared some of the benefits they perceived about teaching online as well as some of the challenges they had faced and challenges they perceived students had faced online. Overall, the teachers feh that there were more similarities than differences in teaching between the two environments, with the main differences being the change from F2F verbal interactions involving body language to online written interactions without body language cues, and the fundamental reliance on technology in the online environment. These findings support previous research in online teaching and learning, and add teachers' perspectives on the factors that stay the same and the factors that change when moving from a F2F environment to an online environment.

Responses of crayfish to a reflective environment /

This thesis examines how dominance status of crayfish alters responses to their own reflection. Crayfish are social animals that fight to develop a dominance hierarchy consisting of dominant and subordinate members. After socialization, crayfish were videotaped in an aquarium with mirrors on one half of the tank and a non-reflective plastic on the other half. Dominants paired for 14 days perform more cornering, turning, crossing and spent more time in a reflective environment versus a non-reflective environment. Subordinate crayfish exhibit more reverse walking in a mirrored environment while isolated crayfish show no preference for reflection. This change in behaviour occurs immediately for dominants paired for 30 min while subordinates require 3 days of pairing to exhibit the same behaviour as subordinate crayfish paired for 14 days. Thus, 30 min of pairing is required to enhance responses to a reflection observed in dominant crayfish while 3 days is required to decrease subordinate responses to a reflection. These findings propose that male socialized crayfish respond to their mirror image as they do a male conspecific. Their responses depend on both their dominance status and the length of socialization which suggests that crayfish are learning to behave in a characteristic manner as a result of their social experience.

Acquisition and transfer of experiential learning from rock climbing to the corporate workplace

There were three purposes to this study. The first purpose was to determine how learning can be influenced by various factors i~ the rock climbing experience. The second purpose was to examine what people can learn from the rock climbing experience. The third purpose was to investigate whether that learning can transfer from the rock climbing experience to the subjects' real life in the workplace. Ninety employees from a financial corporation in the Niagara Region volunteered for this study. All subjects were surveyed throughout a one-day treatment. Ten were purposefully selected one month later for interviews. Ten themes emerged from the subjects in terms of what was learned. Inspiration, motivation, and determination, preparation, goals and limitations, perceptions and expectations, confidence and risk taking, trust and support, teamwork, feedback and encouragement, learning from failure, and finally, skills and flow. All participants were able to transfer what was learned back to the workplace. The results of this study suggested that subjects' learning was influenced by their ability to: take risks in a safe environment, fail without penalty, support each other, plan without time constraints, and enjoy the company of fellow workers that they wouldn't normally associate with. Future directions for research should include different types of treatments such as white water rafting, sky diving, tall ship sailing, or caving.

A study of self-directed learning at an Ontario college of applied arts and technology

Ontario Colleges of Applied Arts and Technology (CAATs) are currently in the process of restructuring to ensure quality, accountability, and accessibility of college education. References to learner involvement and self-directed learning are prevalent. "Alternative delivery" and "paradigm shift" are current buzzwords within the Ontario CAAT system as an environment is created supportive of change. Instability of funding has also dictated a need for change. Therefore, a focus has become quality of learning with less demand on public resources. This qualitative case study was conducted at an Ontario CAAT to gather descriptive, perceptual data from post-secondary community college educators who were identified as supportive of self-directed learning and from post-secondary, traditional-aged college students who were perceived by their educators to be selfdirected learners. This college was selected because of initiatives to modify its academic paradigm to encourage what was reputed in the Ontario CAAT system to be self-directed learning. The purpose of this study was to investigate how postsecondary, traditional-aged college students and their educators perceive self-directed learning as part of the teaching-learning experience within a community college setting. Educator participants of the study were selected based on the results of a teaching and learning survey intended to identify educators supportive of self-directed learning. A total of 317 surveys were distributed to every full-time educator at the sample college; 192 completed surveys were returned for a return rate of 61 %. Of these, 8% indicated instructional beliefs and values supportive of self-directed learning. A purposive sample of six educators was selected using a maximulp variation sampling strategy. A network selection sampling strategy was used to select a purposive sample of seven post-secondary students who were identified by the sample educators as selfdirected learners. The results of the study show that students and educators have similar perspectives and operating definitions of self-directed learning and all participants believe they either practice or facilitate self-directed learning. However, their perspectives and practices are not consistent with the literature which emphasizes learner autonomy or control in course structure and content. A central characteristic of the participants represented in this study is the service-oriented professions with which each is associated. Experientiallearning opportunities were highly valued for the options provided in increasing learner independence and competencies in reflective practice. Although there were discrepancies between espoused theory and theory in practice in terms of course structure, the process of self-directed learning was being practiced and supported outside the classroom structure in clinical settings, labs and related experiences.

Experiential learning instruments : an examination of the concept of the instrumental approach in experiential learning

Experiential Learning Instruments (ELls) are employed to modify the leamer's apprehension and / or comprehension in experiential learning situations, thereby improving the efficiency and effectiveness of those modalities in the learning process. They involve the learner in reciprocally interactive and determining transactions with his/her environment. Experiential Learning Instruments are used to keep experiential learning a process rather than an object. Their use is aimed at the continual refinement of the learner's knowledge and skill. Learning happens as the leamer's awareness, directed by the use of Ells, comes to experience, monitor and then use experiential feedback from living situations in a way that facilitates knmvledge/skill acquisition, self-correction and refinement. The thesis examined the literature relevant to the establishing of a theoretical experiential learning framework within which ELls can be understood. This framework included the concept that some learnings have intrinsic value-knowledge of necessary information-while others have instrumental value-knowledge of how to learn. The Kolb Learning Cycle and Kolb's six characteristics of experiential learning were used in analyzing three ELls from different fields of learning-saxophone tone production, body building and interpersonal communications. The ELls were examined to determine their learning objectives and how they work using experiential learning situations. It was noted that ELls do not transmit information but assist the learner in attending to and comprehending aspects of personal experience. Their function is to telescope the experiential learning process.

Experiences and meaning of the aquatic environment for individuals with physical disabilities

This study examines the experiences and meaning of physical activity in the aquatic environment to enhance social, cultural and political understanding of its impact in the lives of individuals with physical disabilities. Interviews, lived experience descriptions and artifacts present an explanation of the felt sense oftheir bodies as they engage in swimming or scuba diving. 11 Combining written, verbal and visual descriptions generated by informants provides a detailed account of the unique qualities of physical activity in the water for those with physical disabilities. Participants' descriptions highlight that context is an important aspect of physical activity among individuals with physical disabilities through discussion of motility and the role of the lived body. Aspects of the aquatic environment create a setting that facilitates forgetfulness of the lived body's presence. Instructors and participants alike will benefit from learning the difference between the object body and the lived body, listening to the body's voice as they participate in physical activity .

The Family Game: A Parent Education Intervention to Increase Positive Parent-Child Interactions in Parents with Learning Difficulties

Children of parents with learning difficulties (LD) are at risk for a variety of developmental problems including behavioural and psychiatric disorders. However, there are no empirically supported programs to prevent behavioural and psychiatric problems in these children. The purpose of the study was to test the effectiveness of a parenting intervention designed to teach parents with learning difficulties positive child behaviour management strategies. A multiple baseline across skills design was used with two parents, who were taught three skills: 1) clear instructions, 2) recognition of compliance and 3) correction of noncompliance. Training scores improved on each skill and maintained at a 1-month follow-up. Scores on generalization cards were high and showed maintenance, but improvements in parenting skills in the naturalistic environment were low at posttest and follow-up. Increases were seen in child compliance at posttest and 1-month follow-up. Results of pre-post social validity measures were also generally positive.

Adult Learning in the Transformational Environments of the Digital Revolution: Connecting Theory and Practice

The purpose of this study was to investigate the learning preferences and the post-secondary educational experiences of a group of Net-Gen adult learners, aged between 18 and 35, currently working in the knowledge economy workplace, and their assessment of how adequately they were prepared to meet the requirements of the knowledge economy workplace. This study utilized an explanatory mixed-method research design. Participants completed a questionnaire providing information on their self-reported learning style preferences, their use of digital tools for formal and informal learning, their use of digital technologies in postsecondary educational experiences, and their use of digital technologies in their workplace. Four volunteers from the questionnaire respondents were selected to participate in interviews based on the diversity of their experiences in higher education, including digital environments, and the diversity of their knowledge economy workplaces. Data collected from the questionnaire were analyzed for descriptive and demographic statistics, and categorized so that common patterns could be identified from information gathered from the online questionnaire and interviews. Findings based on this study indicated that these Net-Gen adult learners were fluent with all types of digital technologies in collaborative environments, expecting their educational experiences to provide a similar experience. Participants clearly expressed an understanding that digital/collaborative aptitudes are essential to successful employment in the knowledge economy workplace. The findings of this study indicated that the majority of participants felt that their post-secondary educational experiences did not adequately prepare them to meet the expectations of this type of working environment.

Reinforcement Learning Approaches to Power System Scheduling

One major component of power system operation is generation scheduling. The objective of the work is to develop efficient control strategies to the power scheduling problems through Reinforcement Learning approaches. The three important active power scheduling problems are Unit Commitment, Economic Dispatch and Automatic Generation Control. Numerical solution methods proposed for solution of power scheduling are insufficient in handling large and complex systems. Soft Computing methods like Simulated Annealing, Evolutionary Programming etc., are efficient in handling complex cost functions, but find limitation in handling stochastic data existing in a practical system. Also the learning steps are to be repeated for each load demand which increases the computation time.Reinforcement Learning (RL) is a method of learning through interactions with environment. The main advantage of this approach is it does not require a precise mathematical formulation. It can learn either by interacting with the environment or interacting with a simulation model. Several optimization and control problems have been solved through Reinforcement Learning approach. The application of Reinforcement Learning in the field of Power system has been a few. The objective is to introduce and extend Reinforcement Learning approaches for the active power scheduling problems in an implementable manner. The main objectives can be enumerated as:(i) Evolve Reinforcement Learning based solutions to the Unit Commitment Problem.(ii) Find suitable solution strategies through Reinforcement Learning approach for Economic Dispatch. (iii) Extend the Reinforcement Learning solution to Automatic Generation Control with a different perspective. (iv) Check the suitability of the scheduling solutions to one of the existing power systems.First part of the thesis is concerned with the Reinforcement Learning approach to Unit Commitment problem. Unit Commitment Problem is formulated as a multi stage decision process. Q learning solution is developed to obtain the optimwn commitment schedule. Method of state aggregation is used to formulate an efficient solution considering the minimwn up time I down time constraints. The performance of the algorithms are evaluated for different systems and compared with other stochastic methods like Genetic Algorithm.Second stage of the work is concerned with solving Economic Dispatch problem. A simple and straight forward decision making strategy is first proposed in the Learning Automata algorithm. Then to solve the scheduling task of systems with large number of generating units, the problem is formulated as a multi stage decision making task. The solution obtained is extended in order to incorporate the transmission losses in the system. To make the Reinforcement Learning solution more efficient and to handle continuous state space, a fimction approximation strategy is proposed. The performance of the developed algorithms are tested for several standard test cases. Proposed method is compared with other recent methods like Partition Approach Algorithm, Simulated Annealing etc.As the final step of implementing the active power control loops in power system, Automatic Generation Control is also taken into consideration.Reinforcement Learning has already been applied to solve Automatic Generation Control loop. The RL solution is extended to take up the approach of common frequency for all the interconnected areas, more similar to practical systems. Performance of the RL controller is also compared with that of the conventional integral controller.In order to prove the suitability of the proposed methods to practical systems, second plant ofNeyveli Thennal Power Station (NTPS IT) is taken for case study. The perfonnance of the Reinforcement Learning solution is found to be better than the other existing methods, which provide the promising step towards RL based control schemes for practical power industry.Reinforcement Learning is applied to solve the scheduling problems in the power industry and found to give satisfactory perfonnance. Proposed solution provides a scope for getting more profit as the economic schedule is obtained instantaneously. Since Reinforcement Learning method can take the stochastic cost data obtained time to time from a plant, it gives an implementable method. As a further step, with suitable methods to interface with on line data, economic scheduling can be achieved instantaneously in a generation control center. Also power scheduling of systems with different sources such as hydro, thermal etc. can be looked into and Reinforcement Learning solutions can be achieved.

A Reinforcement Learning Algorithm to Economic Dispatch Considering Transmission Losses

Reinforcement Learning (RL) refers to a class of learning algorithms in which learning system learns which action to take in different situations by using a scalar evaluation received from the environment on performing an action. RL has been successfully applied to many multi stage decision making problem (MDP) where in each stage the learning systems decides which action has to be taken. Economic Dispatch (ED) problem is an important scheduling problem in power systems, which decides the amount of generation to be allocated to each generating unit so that the total cost of generation is minimized without violating system constraints. In this paper we formulate economic dispatch problem as a multi stage decision making problem. In this paper, we also develop RL based algorithm to solve the ED problem. The performance of our algorithm is compared with other recent methods. The main advantage of our method is it can learn the schedule for all possible demands simultaneously.

Reinforcement Learning by Policy Search

One objective of artificial intelligence is to model the behavior of an intelligent agent interacting with its environment. The environment's transformations can be modeled as a Markov chain, whose state is partially observable to the agent and affected by its actions; such processes are known as partially observable Markov decision processes (POMDPs). While the environment's dynamics are assumed to obey certain rules, the agent does not know them and must learn. In this dissertation we focus on the agent's adaptation as captured by the reinforcement learning framework. This means learning a policy---a mapping of observations into actions---based on feedback from the environment. The learning can be viewed as browsing a set of policies while evaluating them by trial through interaction with the environment. The set of policies is constrained by the architecture of the agent's controller. POMDPs require a controller to have a memory. We investigate controllers with memory, including controllers with external memory, finite state controllers and distributed controllers for multi-agent systems. For these various controllers we work out the details of the algorithms which learn by ascending the gradient of expected cumulative reinforcement. Building on statistical learning theory and experiment design theory, a policy evaluation algorithm is developed for the case of experience re-use. We address the question of sufficient experience for uniform convergence of policy evaluation and obtain sample complexity bounds for various estimators. Finally, we demonstrate the performance of the proposed algorithms on several domains, the most complex of which is simulated adaptive packet routing in a telecommunication network.