Privileged Learning Research Articles

Learning robust autonomous navigation and locomotion for wheeled-legged robots.

Autonomous wheeled-legged robots have the potential to transform logistics systems, improving operational efficiency and adaptability in urban environments. Navigating urban environments, however, poses unique challenges for robots, necessitating innovative solutions for locomotion and navigation. These challenges include the need for adaptive locomotion across varied terrains and the ability to navigate efficiently around complex dynamic obstacles. This work introduces a fully integrated system comprising adaptive locomotion control, mobility-aware local navigation planning, and large-scale path planning within the city. Using model-free reinforcement learning (RL) techniques and privileged learning, we developed a versatile locomotion controller. This controller achieves efficient and robust locomotion over various rough terrains, facilitated by smooth transitions between walking and driving modes. It is tightly integrated with a learned navigation controller through a hierarchical RL framework, enabling effective navigation through challenging terrain and various obstacles at high speed. Our controllers are integrated into a large-scale urban navigation system and validated by autonomous, kilometer-scale navigation missions conducted in Zurich, Switzerland, and Seville, Spain. These missions demonstrate the system's robustness and adaptability, underscoring the importance of integrated control systems in achieving seamless navigation in complex environments. Our findings support the feasibility of wheeled-legged robots and hierarchical RL for autonomous navigation, with implications for last-mile delivery and beyond.

Weakly privileged learning with knowledge extraction

Learning using privileged information (LUPI) has shown promise in improving supervised learning by embedding additional knowledge. However, its reliance on the assumption of readily available privileged information may not hold true in practical scenarios due to limitations in access or confidentiality. To address these challenges, this paper presents a novel weakly privileged learning (WPL) framework, integrating knowledge extraction methods within the LUPI context. An effective strategy is proposed to implement the WPL framework, where knowledge extraction techniques generate a weight matrix as weak privileged information. Extensive experiments employing various existing knowledge extraction techniques demonstrate that the proposed WPL outperforms traditional supervised learning and approaches the performance of standard privileged learning where privileged information is given in advance. This research establishes WPL as a promising learning paradigm, addressing limitations in privileged information availability and advancing the field of machine learning in practical settings.

Coarse-grained privileged learning for classification

Privileged information, a form of prior knowledge, can significantly enhance traditional machine learning performance through a novel paradigm known as learning using privileged information (LUPI). Although effective, current studies on LUPI require a distinct piece of privileged information per input, and these fine-grained priors are difficult to collect in practice. To this end, this paper proposes a brand new problem of learning with class-wise privileged information, where instances within the same class share identical privileged information. As far as we know, this problem has not yet been explored. We build a support vector machine with coarse-grained class-wise priors (CGSVM+) and put forward a novel and reliable augmenting strategy to solve it. In addition, two datasets are collected from nature reserves in Xinjiang, China, along with their class-wise privileged information annotated by professionals. Extensive experiments demonstrate the effectiveness of CGSVM+, with the best average accuracy of 80.16% (94.70%) and the best average F-score of 79.87% (94.57%) on the plant (animal) datasets.

Pedagogical experiences in non-formal education spaces: stop Motion workshops as methodological contributions to teaching and learning

This text comes from two experiences of learning in non-formal spaces, such as museums and science centers. The first experience is a short course held in 2017 at the Federal University of Viçosa (UFV) during the Farmer's Week. The second experience refers to thematic workshops held in the classroom. The authors in this experience report examine the significance of activities in non-formal learning spaces, in this specific case, museums and science centers, proposing the use of cinematographic arts, with a focus on animations, to integrate museums-schools in the community /university through the meaning and re-signification of museums as privileged learning spaces. As a result, we experience the creation of new methodological instruments in developing new pedagogical practices involving non-formal learning spaces, re-signifying the extra-class knowledge of the investigated subjects.

Don’t think before you speak: on the gradual formation of thoughts during speech

ABSTRACT The common reproach: ‘think before you speak!’, epitomises an educational paradigm in which speech is only the act of transmission finalised ideas. In his inspiring short essay ‘On the Gradual Formation of Thoughts During Speech’ from 1805, the German writer Heinrich von Kleist challenged this approach when he described his habit of talking in order to think. Leaning on cognitive and philosophical theories that support Kleist’s intuitive insights, this study reads his essay as a practical guide in the learning paradigm that I call ‘speak-to-think’. While pedagogical theories have emphasised the importance of interpersonal dialogues and speech in the classroom for thinking and learning, I suggest that self-talk is a paradigmatic manifestation of speak-to-think, and a privileged learning method. Self-talk creates an internal dialogue that can promote meta-cognitive skills and critical thinking, while also challenging the normative perception of speech as a mere conduit for effective communication.

Learning high-speed flight in the wild.

Quadrotors are agile. Unlike most other machines, they can traverse extremely complex environments at high speeds. To date, only expert human pilots have been able to fully exploit their capabilities. Autonomous operation with onboard sensing and computation has been limited to low speeds. State-of-the-art methods generally separate the navigation problem into subtasks: sensing, mapping, and planning. Although this approach has proven successful at low speeds, the separation it builds upon can be problematic for high-speed navigation in cluttered environments. The subtasks are executed sequentially, leading to increased processing latency and a compounding of errors through the pipeline. Here, we propose an end-to-end approach that can autonomously fly quadrotors through complex natural and human-made environments at high speeds with purely onboard sensing and computation. The key principle is to directly map noisy sensory observations to collision-free trajectories in a receding-horizon fashion. This direct mapping drastically reduces processing latency and increases robustness to noisy and incomplete perception. The sensorimotor mapping is performed by a convolutional network that is trained exclusively in simulation via privileged learning: imitating an expert with access to privileged information. By simulating realistic sensor noise, our approach achieves zero-shot transfer from simulation to challenging real-world environments that were never experienced during training: dense forests, snow-covered terrain, derailed trains, and collapsed buildings. Our work demonstrates that end-to-end policies trained in simulation enable high-speed autonomous flight through challenging environments, outperforming traditional obstacle avoidance pipelines.

Educational Inequality Regimes amid Algebra-for-All: The Provision and Allocation of Expanding Educational Opportunities.

Schools can approach the task of sorting students to privileged learning opportunities in different ways, potentially creating distinct and durable educational inequality regimes. We test this idea by exploring variation in socioeconomic inequalities in advanced mathematics course-taking across California middle schools during a statewide algebra-for-all initiative. This case provides unique insight into local stratification processes since the state pressured schools to boost advanced course enrollments but provided little guidance about how to do so. We distinguish two critical organizational processes: the provision of different types of opportunities and the allocation of students to opportunities. The former, we argue, creates the potential for inequality; the latter determines what level of inequality is realized. Using panel data for all public middle schools in the state over a decade, we demonstrate a curvilinear association between opportunities and inequality, with disparities highest when opportunities are most differentiated. However, allocations at most schools were less unequal than would be expected under a test-based meritocratic allocation regime. Further, we find substantial school-level variation which is systematically related to organizational characteristics and consistent over time. These patterns provide evidence for local educational inequality regimes.

Improving Semantic Analysis on Point Clouds via Auxiliary Supervision of Local Geometric Priors.

Existing deep learning algorithms for point cloud analysis mainly concern discovering semantic patterns from the global configuration of local geometries in a supervised learning manner. However, very few explore geometric properties revealing local surface manifolds embedded in 3-D Euclidean space to discriminate semantic classes or object parts as additional supervision signals. This article is the first attempt to propose a unique multitask geometric learning network to improve semantic analysis by auxiliary geometric learning with local shape properties, which can be either generated via physical computation from point clouds themselves as self-supervision signals or provided as privileged information. Owing to explicitly encoding local shape manifolds in favor of semantic analysis, the proposed geometric self-supervised and privileged learning algorithms can achieve superior performance to their backbone baselines and other state-of-the-art methods, which are verified in the experiments on the popular benchmarks.

Serious Game Design: Presenting a New Generic Creative Reflection Framework

Serious games are transcending the way of teaching and learning nowadays. Since serious games balance playful and educational aspects, it constitutes a privileged learning method for the digital natives. However, they are ubiquitous in today's society and their stakes and assets remain largely fuzzy. This paper describes different models and frameworks for serious game design. Between theoretical and technical consideration, authors propose a new reflection model that aims to help pedagogues guiding their conception of serious games. This model is a result of a thematic synthesis analyzing more than 40 texts offering a three-layered model covering the simplest to the most advanced design. Three study cases are announced to be conducted by the end of this year in order to challenge and evaluate this new model.

Iterative Privileged Learning.

While in the learning using privileged information paradigm, privileged information may not be as informative as example features in the context of making accurate label predictions, it may be able to provide some effective comments (e.g., the values of the auxiliary function) like a human teacher on the efficacy of the learned model. In a departure from conventional static manipulations of privileged information within the support vector machine framework, this paper investigates iterative privileged learning within the context of gradient boosted decision trees (GBDTs). As the learned model evolves, the comments learned from privileged information to assess the model should also be actively upgraded instead of remaining static and passive. During the learning phase of the GBDT method, new DTs are discovered to enhance the performance of the model, and iteratively update the comments generated from the privileged information to accurately assess and coach the up-to-date model. The resulting objective function can be efficiently solved within the gradient boosting framework. Experimental results on real-world data sets demonstrate the benefits of studying privileged information in an iterative manner, as well as the effectiveness of the proposed algorithm.

Multi-view transfer learning with privileged learning framework

In this paper, we present a multi-view transfer learning model named Multi-view Transfer Discriminative Model (MTDM) for both image and text classification tasks. Transfer learning, which aims to learn a robust classifier for the target domain using data from a different distribution, has been proved to be effective in many real-world applications. However, most of the existing transfer learning methods map across domain data into a high-dimension space which the distance between domains is closed. This strategy always fails in the multi-view scenario. On the contrary, the multi-view learning methods are also difficult to extend in the transfer learning settings. One of our goals in this paper is to develop a model which can perform better in both multi-view and transfer learning settings. On the one hand, the problem of multi-view is implemented by the paradigm of learning using privileged information (LUPI), which could guarantee the principle of complementary and consensus. On the other hand, the model adequately utilizes the source domain data to build a robust classifier for the target domain. We evaluate our model on both image and text classification tasks and show the effectiveness compared with other baseline approaches.

Desafios na formação do Cirurgião-Dentista para o SUS

A mudança na estrutura curricular realizada pelas escolas de Odontologia, com a inserção de estágios e clínicas integradas, fez com que o desafio da formação profissional para o Sistema Único de Saúde (SUS) fosse possível. Entretanto, ainda existem cursos e docentes que valorizam a Odontologia tecnicista, distanciando os futuros profissionais do atendimento generalista, que é uma das necessidades do SUS. Estas mudanças curriculares possibilitaram um processo de aprendizagem privilegiado, com a promoção e disseminação de saúde por meio de atividades coletivas, acolhimento ampliado em clínica e participação da sociedade em um trabalho de equipe. Neste contexto, a modificação curricular nos cursos de graduação em Odontologia é de extrema necessidade, porém sabe-se que essa mudança deve ser feita de maneira gradativa e constante, garantindo que as mudanças implementadas sejam eficazes e preparem os acadêmicos à sua vida profissional.

Whither social theory?

What is the place of social theory in mathematics education research, and what is it for? This special issue of Educational Studies in Mathematics offers insights on what could be the role of some sociological theories in a field that has historically privileged learning theories coming from psychology and mathematics as the main theoretical frames informing research. Although during the last 10 years the term “socio-cultural” has become part of the accepted and widespread trends of mathematics education research when addressing learning, this issue gathers a collection of papers that depart from a “socio-cultural” approach to learning and rather deploy sociological theories in the analysis of mathematics education practices. In this commentary paper, we will point to what we see to be the contributions of these papers to the field. We will do so by highlighting issues that run through the six papers. We will try to synthetize what we think are the benchmarks of the social approach to mathematics education that they propose. We will also take a critical stance and indicate some possible extensions of the use of social theory that are not addressed in this special issue but nonetheless are worth being explored for a fuller understanding of the “social” in mathematics education.

Afterword

This editorial marks a significant moment for Nursing Inquiry. After over 10 years as editor-in-chief, I am delighted to be handing over to Sally Thorne, professor at University of British Columbia. The editorship of Nursing Inquiry has provided me with a wonderful opportunity for academic growth that I owe entirely to Judith Parker, who skillfully managed to slot me into the driver’s seat before I realized what was happening. Over the years, I have tried to stay true to the brilliant vision of Parker and her co-conspirators and to ensure Nursing Inquiry functions as a forum for new ways of thinking about practice. The importance of such a forum cannot be underestimated. Impact factors and H-index rankings can make us output orientated and risk averse. Moreover, it is increasingly difficult to publish scholarship that is not ‘data-driven’ in the health sciences and therefore Nursing Inquiry plays a critical role in fostering debate, engaging with divergent scholars and supporting the theoretical work that is the foundation of good empirical work. Nursing Inquiry has always maintained the highest production standards, patiently insisted on rigorous reviews from world experts, and worked hard to remain receptive to new ideas and new approaches to practice from scholars around the world. The opportunities to reach this global scholarly community have been enormously enhanced by the expanded e-platforms provided by Wiley-Blackwell and their formidable marketing and distribution network. It has been very exciting as editor to discover the real connection nurses have with Nursing Inquiry in places as diverse as Brazil and Singapore. It is a unique scholarly community – eclectic, creative, risk taking and, at times, iconoclastic. One vital element of our constituency has been doctoral students who have looked to Nursing Inquiry during their struggles with their theory chapters. A great many of our papers eventually come from these very students, as they complete their studies and are then in a position to add to the scholarship that guided them through their doctoral work. The community that supports Nursing Inquiry, whose voice it represents, includes philosophers, anthropologists, historians, political scientists, social theorists, psychoanalysts, ethicists and sociologists. It is this rich interface between nurses, health professionals, the social sciences and the humanities that has made this journal unique. Its conceptualization of nursing scholarship has been broad – defined by the topic of interest rather than the disciplinary field. Although controversial in some quarters, in my view this path has proved its worth as Nursing Inquiry has continued to publish fresh and interesting work, work that makes us think again, over the years. While the editor-in-chief is the public face of the journal, Nursing Inquiry has always been a team effort. The two deputy editors, Michael Traynor in the UK and Judith Parker in Australia, have played a critical role in maintaining the quality and consistency of the journal over the years and I need to thank them for their willingness to pick up whatever I have asked of them. Their advice has always been unerring. Carla Taines, the editorial manager, has been with Nursing Inquiry since its early days and has been a joy to work with. Carla is a remarkable editor, and as the many hundreds of authors who have worked with her over the years will attest, her professionalism and wisdom have provided the backbone to the journal and certainly kept me on track. Many will have interacted with the editorial assistant in their dealings with the journal. I have had the pleasure of working for many years with Debbie Fleming (in Melbourne) and Brigid Nelson (in Toronto). Both were wonderful to work with, very responsive to authors and reviewers (and on occasion caught in the middle) and always highly professional. Colleagues at Blackwell, now Wiley-Blackwell, too, have been enormously committed to the journal’s success and very supportive of the editorial team. Nursing Inquiry is now in great hands. Sally Thorne will be a terrific editor-in-chief and will take the journal and its very special community of scholars exactly where it needs to be. I know it will continue to be a beacon for doctoral students and scholars around the world as they grapple with the ever deepening complexities of practice. My editorship has been a wonderful and privileged learning experience. I thank everyone for their support and look forward to seeing the journal go from strength to strength under Sally’s leadership.

“Gorillas out of the mist”: A pediatric ophthalmologist in the zoo

Charles Darwin explored natural selection during his 4-year trip on the HMS Beagle. From findings of this expedition he constructed his evolution theory. In this Knapp Lecture I would like to take you as a pediatric ophthalmologist on a journey through the Rotterdam Zoo. It will explain why a pediatric ophthalmologist is the perfect eye consultant for a zoo. You will see why an abnormal head position developed in a penguin, anisocoria in a rabbit owl, uveitis and glaucoma as a late complication of a mature cataract in a mouse maki. Retinitis pigmentosa was discovered in a potto. PHPV/PFV is not only seen in humans but also in sand wallaby. Cataract surgery in a Przewalski horse, a snow owl, a codfish, and a Californian sea lion will be demonstrated. Cataract surgery with IOL implantation saved vision and social status in two young lowland gorillas. Working as a pediatric ophthalmologist in a zoo is a privileged learning experience, which gives positive cross-fertilization between animal and human sciences. It is an amazing experience to see how natural selection has given all species the most optimal visual system adapted for their specific habitat.

A model for the enhancement of autonomy

This article presents an innovative model for the enhancement of learner autonomy and shows some of its results and challenges. The model entails investigation of problematic areas of individual students' foreign language learning processes, identification of their privileged learning styles, use of technological tools to improve learning autonomy, development of a wider range of language learning strategies, and implementation of self-monitoring and self-evaluation routines. This model has been applied for the last three years with undergraduate students, earning a B.A. in Modern Languages (English, French or German), who will become foreign language teachers in elementary and high schools in Brazil. Results are threefold: first, the model has proven its efficiency in providing scaffolding for students autonomous language learning; second, autonomy developing experiences lived by subjects who are studying to be teachers can be mirrowed in their future professional lives with their own students; and thirdly, data derived from the participants in the study can shed some light into the variety of ways people learn in the local context.

Deconstructing Diversity Discourses in the Field of Educational Management and Leadership

Discourses of diversity have supplanted those of equal opportunity or social justice in many Western democratic societies. While the notion of diversity is seemingly empowering through its recognition of cultural, religious, racial and gender difference within nation states, the emergence of this discourse during the 1990s has been in the context of neoliberal managerialist discourses that assume social action is fully explicable through theories of maximizing self interest. Thus notions of diversity, while originating in collective demands of social movements of feminism, anti racism and multiculturalism of the 1970s and 1980s, have in recent times privileged learning and leadership as an individual accomplishment and not a collective practice. Thus the dominant discourse of diversity is more in alignment with the deregulatory aspects of the increasingly managerial and market orientation of schooling, decentring earlier discourses of more transformatory notions premised upon reducing inequality and discrimination and developing ‘inclusivity’ in and through schooling. This paper provides a contextual and conceptual framework through which to explore the intersections and divergences of discourses of diversity in schools and their practical application.