Search

Background: Elite athletes often take part in group trainings and use teammates as learning resources. Despite this, research on the training and learning of elite athletes tends to characterise this training and learning as primarily individual.Purpose: This study, explores interrelated learning processes among elite athletes by exploring the performance-related learning that takes place between elite trampoline athletes in their training environment. The case will be made that such learning may be described most accurately as apprenticeship learning.Participants: The research focuses on a case study involving two Danish synchronised trampoline jumpers, Daniel and Peter, and their training as part of the Danish national trampolining team.Data collection: The data were generated through participant observation. During 10 days of observations, the principal researcher held informal talks with the athletes and the coach and took descriptive field notes. At the conclusion of the observation period, each athlete submitted to an individual, semi-structured interview.Data analysis: A theoretical reading of the data was carried out to facilitate interpretations that went beyond observations and the athletes' own descriptions, in order to reach a deeper understanding of how practice facilitates learning.Results: We encircle the athletes' interrelated learning processes by introducing the training environment of the national team and situations in which the athletes guide each other verbally or by jumping together.Discussion: We argue that the practice of the Danish national trampolining team can be considered a community of practice. Taking point of departure in our theoretical perspective we discuss how verbal and tacit bodily exchanges between athletes become opportunities for learning.Conclusion: In a practice containing varied resources for learning, we show that athletes can be each others' performance analysts, guiding each other through the use of metaphors and cues which disclose the practical meaning of how to overcome specific practical challenges. We also show that the athletes can be each others' sparring partners when they perform their sport together. In this form of interaction they directly feel and impact the other's performance, which means that both athletes encounter opportunities for learning.

We consider the problem of imitation learning when the examples, provided by an expert human, are scarce. Apprenticeship learning via inverse reinforcement learning provides an efficient tool for generalizing the examples, based on the assumption that the expert's policy maximizes a value function, which is a linear combination of state and action features. Most apprenticeship learning algorithms use only simple empirical averages of the features in the demonstrations as a statistics of the expert's policy. However, this method is efficient only when the number of examples is sufficiently large to cover most of the states, or the dynamics of the system is nearly deterministic. In this paper, we show that the quality of the learned policies is sensitive to the error in estimating the averages of the features when the dynamics of the system is stochastic. To reduce this error, we introduce two new approaches for bootstrapping the demonstrations by assuming that the expert is near-optimal and the dynamics of the system is known. In the first approach, the expert's examples are used to learn a reward function and to generate furthermore examples from the corresponding optimal policy. The second approach uses a transfer technique, known as graph homomorphism, in order to generalize the expert's actions to unvisited regions of the state space. Empirical results on simulated robot navigation problems show that our approach is able to learn sufficiently good policies from a significantly small number of examples.

The purpose of this research was to examine the consistency between students’ practical and formal understandings of scientific epistemologies (also known as nature of science (NOS) understandings) in the context of a research apprenticeship program. Six high school student participants of a residential summer research apprenticeship program at a major university in the southeastern USA were interviewed twice during their experience to elicit their perspectives regarding their practical epistemologies. A phenomenological approach was used to analyze these interviews. The students held practical epistemological understandings of scientific knowledge that were described as being developmental, valuable, formulaic, and authoritative. A survey administered at the end of the program was used to reveal students’ formal epistemologies of science. These practical and formal epistemologies were described in terms of Sandoval’s (Science Education 89:634–656, 2005) epistemological themes and then compared for all participants. Findings revealed that, for most students, at least some level of consistency was present between their formal and practical epistemological understandings of each theme. In fact, for only one student with one theme, no consistency was evident. These results hold implications for the teaching, learning, and assessment of NOS understandings in these contexts as well as for the design of apprenticeship learning experiences in science.

The project personnel of the VRoma Project have created an on-line virtual environment that offers new possibilities for Latin teachers and students. Through intensive workshops and follow-up mentoring based on apprenticeship learning, the project personnel engage teachers in an egalitarian multilevel community of practice dedicated to using Internet technology to foster the teaching and learning of the Latin language and Roman culture. For students, it provides opportunities not otherwise available for authentic, situated learning of Latin involving the reading and translation of Latin texts in their physical and cultural context, communication in Latin in an imaginatively created native environment, and emulation of expert practitioners through construction of textual commentaries to be "published" within the virtual environment.

The beginning farmer phenomenon offers an array of possibilities for facilitating social, economic, and political changes in the agrifood system. Apprenticeships within both formal and informal institutions are increasingly important in the education and social connectivity of beginning farmers. Although apprenticeship opportunities are popular for "new farmers," "aspiring farmers," and their on-farm hosts for a number of reasons, a critical approach is necessary in the design and nature of these experiences, in light of inequitable structural conditions that may reproduce potentially insurmountable barriers to new farm entry and sustainability. Drawing upon alternative agrifood movement discourse and social reproduction at work within critical traditions of sociocultural learning, we illustrate on-farm apprenticeship learning from a critical perspective in order to better describe and understand this form of beginning farmer education. We share findings from a mixed-methods empirical study of on-farm apprenticeship learning in Virginia, where we focus on the practices, structures, and institutional activity that inform on-farm apprenticeship experiences. This study sought to answer the questions: what kinds of on-farm apprenticeships are available, to whom, and in what ways? Also, what are important educational practices, structures, and/or institutions that support on-farm apprenticeship learning? Data are derived from qualitative interviews of host farmer/educators, on-farm apprentices, and new farmers who were recently apprentices; and from a quantitative survey of Virginia farmers who host apprentices. Our findings situate on-farm apprenticeship within a broader discourse about farm labor, as we open the discussion surrounding the relationship between difficulties experienced by small, diversified farms in meeting their labor needs, and the growing popularity of the apprenticeship model on individual farms. We also explore how cultural whiteness within alternative agrifood movements (AAMs) translates to low inclusivity of historically underrepresented groups, and consider how the low- or no-pay model for the tenured duration of the apprenticeship may affect structural barriers to entry for members of low socioeconomic groups, within on-farm apprenticeship and thus within beginning farmer education. Through the themes that emerged in our study, we posit considerations for social justice implications of on-farm apprenticeship, offer several recommendations for the practice and planning of on-farm apprenticeship, and lay groundwork for future exploration of the ways in which the apprenticeship model may reproduce equitable learning spaces.

Simulation Balancing is an optimization algorithm to automatically tune the parameters of a playout policy used inside a Monte Carlo Tree Search. The algorithm fits a policy so that the expected result of a policy matches given target values of the training set. Up to now it has been successfully applied to Computer Go on small 9 × 9 boards but failed for larger board sizes like 19 × 19. On these large boards apprenticeship learning, which fits a policy so that it closely follows an expert, continues to be the algorithm of choice. In this paper we introduce several improvements to the original simulation balancing algorithm and test their effectiveness in Computer Go. The proposed additions remove the necessity to generate target values by deep searches, optimize faster and make the algorithm less prone to overfitting. The experiments show that simulation balancing improves the playing strength of a Go program using apprenticeship learning by more than 200 ELO on the large board size 19 × 19.

Multi-Objective Reinforcement Learning (MORL) can be divided into two approaches according to the number of acquired policies. One approach learns a single policy that makes the agent reach a single arbitral Pareto optimal solution, and the other approach learns multiple policies that correspond to each Pareto optimal solution. The latter approach finds the multiple policies simultaneously, however, it incurs significant computational cost. In many real-world cases, learning a single solution is sufficient in the multi-objective context. In this paper, we focus on the former approach where a suitable weight of each object must be defined. To estimate the weight of each object as parameters, we utilize Q-values on the expert's trajectory, which indicates the optimal sequence of actions. This approach is an analogy obtained from apprenticeship learning via inverse reinforcement learning. We evaluate the proposed method using a well-known MORL benchmark problem, i.e., the Deep Sea Treasure environment.

Several studies have examined how doctors learn in the workplace, but research is needed linking workplace learning with the organisation of doctors' daily work. This study examined residents' and consultants' attitudes and beliefs regarding workplace learning and contextual and organisational factors influencing the organisation and planning of medical specialist training. An explorative case study in three paediatric departments in Denmark including 9 days of field observations and focus group interviews with 9 consultants responsible for medical education and 16 residents. The study aimed to identify factors in work organisation facilitating and hindering residents' learning. Data were coded through an iterative process guided by thematic analysis. Findings illustrate three main themes: (1) Learning beliefs about patient care and apprenticeship learning as inseparable in medical practice. Beliefs about training and patient care expressed in terms of training versus production caused a potential conflict. (2) Learning context. Continuity over time in tasks and care for patients is important, but continuity is challenged by the organisation of daily work routines. (3) Organisational culture and regulations were found to be encouraging as well inhibiting to a successful organisation of the work in regards to learning. Our findings stress the importance of consultants' and residents' beliefs about workplace learning as these agents handle the potential conflict between patient care and training of health professionals. The structuring of daily work tasks is a key factor in workplace learning as is an understanding of underlying relations and organisational culture in the clinical departments.

This paper considers the problem of learning task specific web-service descriptions from traces of users successfully completing a task. Unlike prior approaches, we take a traditional machine-learning perspective to the construction of web-service models from data. Our representation models both syntactic features of web-service schemas including lists and optional elements, as well as semantic relations between objects in the task. Together, these learned models form a full schematic model of the dataflow. Our theoretical results, which are the main novelty in the paper, show that this structure can be learned efficiently: the number of traces required for learning grows polynomially with the size of the task. We also present real-world task descriptions mined from tasks using online services from Amazon and Google.

© 2015 IEEE. In this paper we introduce a novel approach to agent control in competitive video games which combines Monte Carlo Tree Search (MCTS) and Apprenticeship Learning (AL). More specifically, an opponent model created through AL is used during the expansion phase of the Upper Confidence Bounds for Trees (UCT) variant of MCTS. We show how this approach can be applied to a game of Capture the Flag (CTF), an environment which is both non-deterministic and partially observable. The performance gain of a controller utilizing an opponent model learned via AL when compared to a controller using just UCT is shown both with win/loss ratios and True Skill rankings. Additionally, we build on previous findings by providing evidence of a bias towards a particular style of play in the AI Sandbox CTF environment. We believe that the approach highlighted here can be extended to a wider range of games other than just CTF.

In this article, it will be argued that in a postmodern society, there is a growing interest in apprenticeship. In a postmodern society, one of the issues dominating work life today is the need to learn local and pragmatic kinds of knowledge due to specialized, rapidly changing, and flexible forms of production. The resurrected interest in apprenticeships that we are witnessing is not due to mere chance or personal contributions to the field; additionally, they are not only a reflection of educational fads. Rather, apprenticeship learning is in line with current postmodern conceptions of knowledge. The change in production involves a relocation of knowledge, the importance and recognition of exemplary forms of acquiring knowledge, the primacy of pragmatic knowledge, knowledge as socially distributed, and knowledge as flexible, changing.

Evaluation of programs that address the lingering effects of human rights abuses during times of conflict is necessary to improve program sustainability and create a knowledge bank about the effectiveness of strategies. Outcomes, however, are hard to measure. Evaluators have to gain understanding of the roots of a conflict, surrounding events, histories, and cultures. Discussed is the concept of culturally responsive evaluation (CRE). A pipeline program, which supports graduate students from traditionally underrepresented population groups in acquiring CRE skills through apprenticeship learning, seminars, and mentorship, is described. The work of 2 program participants, who evaluated programs—1 serving survivors of torture and the other children of refugees—are given as examples.

This article explores teaching assistants' (TAs) perceptions of suggestions that English schools should encourage large numbers of ‘low‐achieving’ school leavers to become apprentice teaching assistants. It reports on a piece of qualitative research that aimed to capture teaching assistants' views of these proposals. Semi‐structured interviews were utilised to enable 44 TAs to express their views. These individuals were all enrolled on a foundation degree programme at a new university in the North of England. The findings of this research suggest that established teaching assistants' reactions to apprenticeships are likely to promote a learning environment where an already restricted form of apprenticeship experience will be subject to further constraint. This article highlights the risks that vocational education and training (VET) of this type poses in terms of reproducing gender and class‐based inequalities. It further maintains that diluting the meaning of apprenticeship learning to meet ambitious gov...

We present a method for inferring the behavior styles of character controllers from a small set of examples. We show that a rich set of behavior variations can be captured by determining the appropriate reward function in the reinforcement learning framework, and show that the discovered reward function can be applied to different environments and scenarios. We also introduce a new algorithm to recover the unknown reward function that improves over the original apprenticeship learning algorithm. We show that the reward function representing a behavior style can be applied to a variety of different tasks, while still preserving the key features of the style present in the given examples. We describe an adaptive process where an author can, with just a few additional examples, refine the behavior so that it has better generalization properties.

Autonomous helicopter flight is widely regarded to be a highly challenging control problem. Despite this fact, human experts can reliably fly helicopters through a wide range of maneuvers, including aerobatic maneuvers at the edge of the helicopter’s capabilities. We present apprenticeship learning algorithms, which leverage expert demonstrations to efficiently learn good controllers for tasks being demonstrated by an expert. These apprenticeship learning algorithms have enabled us to significantly extend the state of the art in autonomous helicopter aerobatics. Our experimental results include the first autonomous execution of a wide range of maneuvers, including but not limited to in-place flips, in-place rolls, loops and hurricanes, and even auto-rotation landings, chaos and tic-tocs, which only exceptional human pilots can perform. Our results also include complete airshows, which require autonomous transitions between many of these maneuvers. Our controllers perform as well as, and often even better than, our expert pilot.

Autonomous helicopter flight is widely regarded to be a highly challenging control problem. As helicopters are highly unstable and exhibit complicated dynamical behavior, it is particularly difficult to design controllers that achieve high performance over a broad flight regime. While these aircraft are notoriously difficult to control, there are expert human pilots who are nonetheless capable of demonstrating a wide variety of maneuvers, including aerobatic maneuvers at the edge of the helicopter's performance envelope. In this paper, we present algorithms for modeling and control that leverage these demonstrations to build high-performance control systems for autonomous helicopters. More specifically, we detail our experiences with the Stanford Autonomous Helicopter, which is now capable of extreme aerobatic flight meeting or exceeding the performance of our own expert pilot.

In The Perception of the Environment (2000), Ingold has argued that differences in cultural knowledge are more a matter of variation in embodied skills than in discursive knowledge. These skills develop through the practitioners' engagement with their environment and in situated social relationships. In order to `discover' for themselves what is taken for granted for experienced practitioners, they have to `fine-tune' their perception through observation and imitation. But how do observations and imitations of others' movements actually transfer into shifts in one's own perception? In her book Loving Nature: Towards an Ecology of Emotion (2002), Milton argued that emotion acts as a learning mechanism to filter attention. I propose that when one observes and imitates in a process of learning, one enters into an empathic relationship with a skilled practitioner. Through synchronization of intentions and movements, emotions spread over and change the practitioners' perception accordingly.

Apprenticeship learning based on inconsistent demonstrations is presented in this paper. We address a problem where given demonstrations are not directly applicable to reward function estimation due to the non-stationarity of an environment or the difference between the dynamics of a robot and a demonstrator. A basic idea of the proposed method is to use a subset of the trajectories sampled from the baseline policy as training data for inverse reinforcement learning. All consistent sample trajectories and inconsistent demonstrations are abstracted by an affine transformation invariant feature. Using the feature, the importance of each sample trajectory is estimated. Rating the sample trajectories based on importance, the training data for inverse reinforcement learning are identified. The validity of our approach is verified through simulation in two scenarios: inconsistency caused by variation of an environment and performance of a robot.

International audience; A popular approach to apprenticeship learning (AL) is to formulate itas an inverse reinforcement learning (IRL) problem. The MaxEnt-IRL algorithm successfully integrates the maximum entropy principleinto IRL and unlike its predecessors, it resolves theambiguity arising from the fact that a possibly large number of policies couldmatch the expert's behavior. In this paper, we study an AL setting in which inaddition to the expert's trajectories,a number of unsupervised trajectories is available. We introduce MESSI,a novel algorithm that combines MaxEnt-IRLwith principles coming from semi-supervised learning. In particular, MESSIintegrates the unsupervised data intothe MaxEnt-IRL framework using a pairwise penalty on trajectories. Empiricalresults in a highway driving and grid-world problems indicate that MESSI is able to take advantage of the unsupervised trajectories and improve the performance ofMaxEnt-IRL.

Reinforcement learning agents can acquire the optimal policy to achieve their objectives based on trials and errors. An appropriate design of reward function is essential, because there are variety of reward functions for the same objective whereas different reward functions would give rise to different learning processes. There is no systematic way to determine a good reward function for a given environment and objective. One possible way is finding a reward function to imitate the learning strategy of a reference agent which is intelligent enough to efficiently adapt even variable environments. In this study, we extended the apprenticeship learning framework in order to imitate a learning reference agent, whose policy may change on the process of optimization. For the imitation above, we propose a new inverse reinforcement learning based on that agent’s history of states and actions. When mimicking a reference agent that was trained with a simple 2-state Markov decision process, the proposed method showed better performance than that by the apprenticeship learning.

Apprenticeship learning occurs via observations while an expert is in action. A hyper-heuristic is a search method or a learning mechanism that controls a set of low level heuristics or combines different heuristic components to generate heuristics for solving a given computationally hard problem. In this study, we investigate into a novel apprenticeship-learning-based approach which is used to automatically generate a hyper-heuristic for vehicle routing. This approach itself can be considered as a hyper-heuristic which operates in a train and test fashion. A state-of-the-art hyper-heuristic is chosen as an expert which is the winner of a previous hyper-heuristic competition. Trained on small vehicle routing instances, the learning approach yields various classifiers, each capturing different actions that the expert hyper-heuristic performs during the search process. Those classifiers are then used to produce a hyper-heuristic which is potentially capable of generalizing the actions of the expert hyper-heuristic while solving the unseen instances. The experimental results on vehicle routing using the Hyper-heuristic Flexible (HyFlex) framework shows that the apprenticeship-learning-based hyper-heuristic delivers an outstanding performance when compared to the expert and some other previously proposed hyper-heuristics.

Narrative is an interactive communicative process occurring in particular social contexts and invoking other texts, tellings, and ideas, including those historically and geographically distant from the immediate context. Telling narratives is part of a larger process of the circulation of knowledge, and the ethnographic study of narrative requires attention to a cultural group’s communicative repertoire. Taking the example of narratives told by Italian artisan stone carvers, this essay explores life history narrative, community memory, and informal pedagogies. The pedagogical practices of informal, apprenticeship learning are rarely recorded in text books but instead are transmitted through narratives that convey not only instruction in particular skills but also values about ways of learning and ways of practicing a skill, including remembered skills no longer practiced. Occurring at the interface between technology and pedagogy, informal learning is deeply embedded in cultural and cross-cultural circulations of knowledge. In a discussion of the uses of narrative in research, including longitudinal ethnographic research projects, the essay addresses central dimensions of narrative research including positionality, frame, local and grand narrative, intertexuality, and dialogic narration.

This article considers teaching and learning as a collaborative enterprise in the workplace. The empirical data have been extracted from a field study among apprentices engaged in electromechanical vocational training and education in a major Danish industrial company. Typically, studies of apprenticeship learning do not view aspects of teaching as decisive parts of workplace learning. In this article, I will pin down the specific character of collaborative teaching and learning in the workplace, and compare this with the features of a teaching situation in school with a teacher transmitting knowledge to the pupils. While instruction at school is organised as part of a specifically planned teaching situation, the instruction in the workplace is a more loosely organised part of the joint activity structure of the daily work tasks. The conclusions reached in the article should be useful for the conceptual understanding of instruction and teaching in workplace learning, as well as for policy makers and professionals responsible for the design of apprenticeships and workplace learning.

Senior house officer schools in histopathology have been established in Leeds, Leicester and Southampton, each training six new recruits each year. Nine hours of protected teaching time is provided each week giving a ratio of apprenticeship learning to formal teaching of 3:1. Evaluations have been very positive. Much of this success is attributed to careful planning and adequate funding. This may be a useful model for other specialties to follow.

This paper describes student learning within the environment of an HFOSS project that is jointly shared between the GNOME Accessibility Team and three academic institutions. This effort differs from many project-based learning efforts in that the project is shared between the academic participants and the HFOSS community. By involving students in an HFOSS project, learning is started via apprenticeship which allows students to learn from professionals while preparing them for their professional life. Learning within the community of an ongoing FOSS project guides students in the first steps towards understanding the importance of life-long learning as well as providing an initial understanding of the ways in which such learning occurs. The results of a student survey and observations of student reflection papers are discussed.

This paper proposes a novel model-free inverse reinforcement learning method based on density ratio estimation under the framework of Dynamic Policy Programming. We show that the logarithm of the ratio between the optimal policy and the baseline policy is represented by the state-dependent cost and the value function. Our proposal is to use density ratio estimation methods to estimate the density ratio of policies and the least squares method with regularization to estimate the state- dependent cost and the value function that satisfies the relation. Our method can avoid computing the integral such as evaluating the partition function. A simple numerical simulation of a grid world navigation, a car driving, and a pendulum swing-up shows its superiority over conventional methods. I. INTRODUCTION Reinforcement Learning (RL) is a computational framework for investigating decision-making processes of both biological and artificial systems that can learn an optimal policy by interacting with an environment (1). There exist several open questions in RL, and one of the critical problems is how we design and prepare an appropriate reward/cost function. It is easy to design a sparse reward function which gives a positive reward when the task is accomplished and zero otherwise, but that makes it hard to find an optimal policy. In some situations, it is easier to prepare examples of a desired behavior than to handcraft an appropriate reward/cost function. Recently, several methods of Inverse Reinforcement Learning (IRL) (2) and apprenticeship learning (3) have been proposed in order to derive a reward/cost function from demonstrator's behaviors. IRL provides a powerful way to implement imitation learning, and it is also a promising ap- proach for understanding the learning processes of biological systems because the reward/cost function specifies the goal of the behavior. However, most of the existing studies (3), (4) require a routine to solve forward reinforcement learning problems with estimated reward/cost functions. This process is usually very time consuming even when the model of the environment is available. Recently, IRL algorithms without solving forward reinforce- ment learning are proposed. The key idea of these methods is to introduce the Kullback-Leibler (KL) divergence. For exam- ple, OptV (5) is the model-based IRL which is derived from the concept of Linearly solvable Markov Decision Process (LMDP) (6). LMDP is a sub-class of Markov Decision Process

Imitating human-like behaviour in action games is a challenging but intriguing task in Artificial Intelligence research, with various strategies being employed to solve the human-like imitation problem. In this research we consider learning human-like behaviour via Markov decision processes without being explicitly given a reward function, and learning to perform the task by observing expert's demonstration. Individual players often have characteristic styles when playing the game, and this method attempts to find the behaviours which make them unique. During play sessions of Super Mario we calculate player's behaviour policies and reward functions by applying inverse reinforcement learning to the player's actions in game. We conduct an online questionnaire which displays two video clips, where one is played by a human expert and the other is played by the designed controller based on the player's policy. We demonstrate that by using apprenticeship learning via Inverse Reinforcement Learning, we are able to get an optimal policy which yields performance close to that of an human expert playing the game, at least under specific conditions.

Apprenticeship experiences have existed for centuries asways to gradually introduce individuals to professionaltrades, skills, and roles. Cobblers, blacksmiths, attorneys,artists, healers, and surgeons have all learned as appren-tices. Traditional apprenticeships focus on specificmethods for carrying out physical skills instrumental toaccomplishing meaningful practical tasks (e.g., buildinga house). Apprenticeship learning occurs through a com-bination of observing, coaching, and practice. The cogni-tive apprenticeship model, initially described by Collins[1] and elaborated by Stalmeijer et al. [2, 3], shares thetraditional apprenticeship’s focus on learning complextasks from experts, but further emphasizes cognitive skillsof the teachers and learners, not as readily observable [1,2]. Cognitive psychotherapy skills refer to clinical reason-ing, the development of a differential diagnosis, and caseformulation. Cognitive skills also include the ability touse emotions, behaviors, precise therapeutic language,and the strategies, tactics, and interventions needed toconduct psychotherapy. Developing psychotherapeuticcognitive skills requires that both supervisor and residentexternalize or think out loud, sharing their thoughts, emo-tional process, and the rationales for their therapeuticchoices. The problems and tasks that supervisors addressare chosen to illustrate the effectiveness of a psychother-apeutic task, discuss mistakes, or demonstrate psychother-apeutic interventions so that psychiatric residents canpractice and improve their reflective psychological andemotion-focused skills.MethodsThe “Apprenticeship Model” is utilized as part of ourPsychotherapy Scholar’s Track (PST) described elsewhere indetail [4]. This track emphasizes use of the common factorsinfluencing psychotherapy outcomes. Residents learn a mini-mum of five psychotherapeutic modalities, including our brieftreatment model [5], psychodynamic, cognitive and behavioraltherapies, family, and group psychotherapies. We consistentlyuse symptom and functional rating scales at the beginning andendofthetreatmentstohelpresidentsquantifypatientoutcomes.Supervisors are a broadly trained scholarly group of psy-chiatrists and psychologists, having 8–45 years of experienceaspsychotherapists.Weareallsalariedpreceptors,workinginour university outpatient department, conducting apprentice-ship cases carved out of our preceptor time.Since no objective measures of psychotherapy supervi-sion competency have ever been consistently validated,we rely on faculty who by virtue of experience, tempera-ment, and reputation are thought to have the necessary“wisdom” and capabilities to teach the skills required toconduct several different forms of psychotherapy.“Wisdom” as used here, is defined by Sternberg [6, 7]as a clinician who has“a deeper understanding of reality,used toward a common good, encompassing the definingcharacteristics of concern for others, psychological under-standing of others, capacity for self-knowledge, the abilityto reframe information, the ability to take the long view ofproblems, readily admitting to and learning from mis-takes, maturity, intuition, and the ability to“see through”situations and circumstances. Our “wise” [6, 7] supervi-sors emphasize common factors affecting psychotherapyoutcomes and psychotherapeutic principles derived from avariety of theoretical perspectives.

Beyond the institution of apprenticeship, most strongly associated with Medieval guilds, apprenticeship can represent a powerful model to think about learning. Through a discussion of some of the most influential theories on apprenticeship learning in the social sciences, this chapter aims to find ways to think about apprenticeship in its complexity and diversity as a mode of learning, and specifically in relation to professional learning.

The purpose of this paper is to synthesize literature related to apprenticeship learning, the sociology of science, and K-12 science education to develop a set of characteristics for designing/evaluating participatory science learning experiences. Following this discussion, we further clarify and illuminate the value of these characteristics for science educators by using them as evaluative criteria for characterizing the experiences of 24 middle school learners who embarked on a 2-week long camp with ''real'' scientists engaged in ''real'' research. We also describe how middle school science teachers supported both reflection- in-practice and reflection-on-practice during the camp, and how an electronic notebook was also leveraged to support both types of reflection. Implications of these characteristics for science education more generally are discussed. fl 2000 John Wiley & Sons, Inc. J Res Sci Teach 38: 70 - 102, 2001

Considering that expert's demonstrations are usually sub optimal and failed demonstrations often have some useful guidance, in this paper, a Discriminative Apprenticeship Learning algorithm is proposed, where the apprentice is taught with the join of failed attempts to acquire the ability that could discriminate the preference and non-preference cases so that to actively take a corresponding action. Since robot usually encounters changing environments, generalization ability is taken into account in the algorithm through which the reward function is recovered under the evaluation of generalization error. The problem of the representation error is also analyzed and involved in the algorithm. To ensure performance of the algorithm, theoretical guarantee is presented. Experiments on a simple car-driving robot and the comparison with a variety of inverse reinforcement learning methods are performed, which illustrate the proposed method is an effective and promising alternative.

An ontogenetically-informed epistemology is necessary to understandings of apprenticeship learning. The methodology required in this enterprise is a constructive phenomenology, a phenomenology that takes into account the fact that as infants, we were apprentices of our own bodies: we all learned our bodies and learned to move ourselves. The major focus of this essay is on infant social relationships that develop on the ground of our original corporeal-kinetic apprenticeship. It shows how joint attention, imitation, and turn-taking - all richly examined areas in infant social development - are the foundation of apprenticeship learning in later adult life. The relationship between each infant capacity and later apprenticeship learning is demonstrated in examples from present-day research, specifically, research in the areas of medicine, sport, music, and tailoring, and research carried out by philosophers on apprenticeship learning.

Against a background of the theoretical basis for the contextual approach to medical education, this paper examines and supports the changes that are occurring in undergraduate medical education throughout the world, before putting up for discussion the suggestion that the changes have not gone far enough. Consideration is given to a model of apprenticeship learning within undergraduate medical education, to the benefits it may offer, and to some of the challenges inherent in its implementation.

The thrust of peer apprenticeship learning is in how an individual's personal beliefs, disposition, and epistemology can be transformed through an “apprenticeship-like” learning process. This paper describes the peer apprenticeship learning situation between two students, Dom (age: 13) and Ming (age: 14), in distributed computer-mediated co-construction of mathematical meanings. Ming was initially procedural or rule-based in his problem solving methods, encountering numerous difficulties with his approaches. Dom, however, was constantly engaged in playing with ideas through conjecturing and perceiving mathematical relationships. In the initial stages of co-construction efforts, Dom and Ming were solving the problems rather independently as each appropriated a radically different epistemology of mathematics – Ming was rule-based whereas Dom was idea-based. In the cause of increasing difficulties, Ming soon recognized that his methods were inadequate and, after a considerable struggle, positioned himself in an “apprentice-like” manner in order to follow Dom's conceptualizations. Through monitoring of Dom's conceptualizations and with personal experimentations to concretize his understanding, Ming was gradually able to assimilate Dom's mathematical meaning perspectives. We depicted such a learning situation as peer apprenticeship learning. As a result of assimilating the disposition towards playing with ideas, Dom and Ming were able to engage in meaningful idea-based social constructivism.

An apprenticeship-learning-based technique is used as a hyper-heuristic to generate heuristics for an online combinatorial problem. It observes and learns from the actions of a known-expert heuristic on small instances, but has the advantage of producing a general heuristic that works well on other larger instances. Specifically, we generate heuristic policies for online bin packing problem by using expert near-optimal policies produced by a hyper-heuristic on small instances, where learning is fast. The "expert" is a policy matrix that defines an index policy, and the apprenticeship learning is based on observation of the action of the expert policy together with a range of features of the bin being considered, and then applying a k-means classification. We show that the generated policy often performs better than the standard best-fit heuristic even when applied to instances much larger than the training set.

Innovation in clinical learning is reported as being useful in preparing health and social care students for clinical patient care, but historically these learners have relied on traditional pedagogies including didactic classroom learning and apprenticeship „practice‟ on live patients in a clinical environment. This paper investigates whether students find it useful to augment traditional learning methods with simulation and video (hybrid learning) as a pre-junct to learning in clinical placement. Replacing the usual initial clinical placement with a 12-week study block employing hybrid-teaching techniques reformed the traditional curriculum for Operating Department Practitioner students. An interpretative phenomenographic methodology was adopted for this study and data was collected through anonymous focus group interviews. The data support two concepts that 1) confidence and self-belief perpetuate the desire for new learning and 2) multi-professional learning develops a professional approach in terms of communication, care intervention and thinking processes. The data presented was obtained using a qualitative phenomenographic approach and the results infer specific advantages of hybrid learning to the participants to supplement traditional teaching methods by addressing theoretical limitations of learning and inequity of placement experiential learning. Limitations to this study are the absence of a control group to directly compare against apprenticeship learning methods alone and the sample group being single site, single cohort. © 2013 GESDAV

Autonomous valve turning is an extremely challenging task for an Autonomous Underwater Vehicle (AUV). To resolve this challenge, this paper proposes a set of different computational techniques integrated in a three-layer hierarchical scheme. Each layer realizes specific subtasks to improve the persistent autonomy of the system. In the first layer, the robot acquires the motor skills of approaching and grasping the valve by kinesthetic teaching. A Reactive Fuzzy Decision Maker (RFDM) is devised in the second layer which reacts to the relative movement between the valve and the AUV, and alters the robots movement accordingly. Apprenticeship learning method, implemented in the third layer, performs tuning of the RFDM based on expert knowledge. Although the long-term goal is to perform the valve turning task on a real AUV, as a first step the proposed approach is tested in a laboratory environment. © 2013 IEEE.

In visual category recognition there is often a trade-off between fast and powerful classifiers. Complex models often have superior performance to simple ones but are computationally too expensive for many applications. At the same time the performance of simple classifiers is not necessarily limited only by their flexibility but also by the amount of labelled data available for training. We propose a semi-supervised wrapper algorithm named apprenticeship learning, which leverages the strength of slow but powerful classification methods to improve the performance of simpler methods. The powerful classifier parses a large pool of unlabelled data, labelling positive examples to extend the dataset of the simple classifier. We demonstrate apprenticeship learning and its effectiveness by performing experiments on the VOC2007 dataset - one experiment improving detection performance on VOC2007, and one domain adaptation experiment, where the VOC2007 classifier is adapted to a new dataset, collected using a GoPro camera.

International audience; This paper provides a comparative study between Inverse Reinforcement Learning (IRL) and Apprenticeship Learning (AL). IRL and AL are two frameworks, using Markov Decision Processes (MDP), which are used for the imitation learning problem where an agent tries to learn from demonstrations of an expert. In the AL Framework, the agent tries to learn the expert policy whereas in the IRL Framework, the agent tries to learn a reward which can explain the behavior of the expert. This reward is then optimized to imitate the expert. One can wonder if it is worth estimating such a reward, or if estimating a Policy is sufficient. This quite natural question has not really been addressed in the literature right now. We provide partial answers, both from a theoretical and empirical point of view.

The growing complexity of contemporary engineering designs requires the use of sophisticated computer-based design tools. Such tools increase productivity in drafting, configuration and calculations. However, the current generation of design tools plays a rather passive role in the entire design process. Recent progress in knowledge-based engineering design, machine learning and knowledge acquisition allows the development of knowledge-based design assistants which could behave as active partners to human designers, rather than as passive graphical or computational tools. This paper presents the shared expertise model (SEM) of interaction between a human designer and a knowledge-based design assistant, in which the design assistant behaves as an apprentice and a collaborator in the design process. The human designer and his computer-based assistant create designs together, with the assistant proposing routine or even innovative designs, and the human designer correcting and finalizing these design, as well as specifying creative designs. In this process, the assistant also learns from the human designer, constantly extending and improving its knowledge base, and becoming a better design assistant. This is achieved by employing apprenticeship multistrategy learning based on a plausible version space representation. Within the SEM framework, a human designer with limited programming capabilities can directly develop and maintain a personalized knowledge-based design assistant. The use of SEM leads to increased capabilities of the expert-assistant design team. In this paper, some of the main features of the SEM model, as well as its implementation with the Disciple toolkit, are illustrated within the domain of computer workstation configuration.

Knowledge elicitation from experts and empirical machine learning are two distinct approaches to knowledge acquisition with differing and mutually complementary capabilities. Learning apprentices have provided environments in which a knowledge engineer may collaborate with a machine learning system allowing for a synergy between the complementary approaches. The Knowledge Factory is a knowledge acquisition environment that allows a domain expert to collaborate directly with a machine learning system without the need for assistance from a knowledge engineer. This requires a different form of environment to the learning apprentice. The paper describes techniques for supporting such interactions and their implementation in a knowledge acquisition environment called The Knowledge Factory.

This paper presents the unified approach to machine learning and knowledge acquisition in the Case-Oriented Expert (COEx) system architecture. Knowledge acquisition techniques are used to delineate a problem class hierarchy for complex real-world planning tasks. This hierarchy is stepwise formalized into a KL-ONE-like representation. Explanation-based abstraction is then applied to previously solved cases, and procedure schemata are constructed for the different problem classes. COEx is a type of apprenticeship learning system where reversed engineering is applied to success cases from the real world. An evaluation of this unification reveals its conceptual soundness and several technical problems.

Given the cultural learning models, an important question to consider is how these cultural models may influence the actual learning process in which the members of each culture engage. I attempt to address this question in this chapter. Before proceeding, however, it is important to lay out different kinds of human learning and clarify to what kind of learning the cultural models are most relevant. KINDS OF HUMAN LEARNING It is a well-established scientific fact that human beings are capable of learning virtually unlimited things. Human learning takes place in many ways, and it starts even before birth. For the most part, infants and young children learn much about the world without being specifically taught; for example, they acquire vocabulary of their native tongue and social norms of behavior without deliberate adult instructions. However, because it proceeds more or less without the need for effort on the part of young children or deliberate instructions on the part of adults, this kind of individual learning is not directly influenced by culture. It is instead testimony to the human capacity to learn.

We present an algorithmic approach for integrated learning and planning in predictive representations. The approach extends earlier work on predictive state representations to the case of online exploration, by allowing exploration of the domain to proceed in a goal-directed fashion and thus be more efficient. Our algorithm interleaves online learning of the models, with estimation of the value function. The framework is applicable to a variety of important learning problems, including scenarios such as apprenticeship learning, model customization, and decision-making in non-stationary domains.

We propose a graph-based algorithm for apprenticeship learning when the reward features are noisy. Previous apprenticeship learning techniques learn a reward function by using only local state features. This can be a limitation in practice, as often some features are misspecified or subject to measurement noise. Our graphical framework, inspired from the work on Markov Random Fields, allows to alleviate this problem by propagating information between states, and rewarding policies that choose similar actions in adjacent states. We demonstrate the advantage of the proposed approach on grid-world navigation problems, and on the problem of teaching a robot to grasp novel objects in simulation.