Fostering autonomous learning by using a workbook to improve reading and writing skills in students of basic level at ASE

Tesis (Licenciado en Lenguas Castellana, Inglés y Francés).--Universidad de La Salle. Facultad de Ciencias de La Educación. Licenciatura en Lengua Castellana, Inglés y Francés, 2014

From vision to actions: Towards adaptive and autonomous humanoid robots

Although robotics research has seen advances over the last decades robots are still not in widespread use outside industrial applications. Yet a range of proposed scenarios have robots working together, helping and coexisting with humans in daily life. In all these a clear need to deal with a more unstructured, changing environment arises. I herein present a system that aims to overcome the limitations of highly complex robotic systems, in terms of autonomy and adaptation. The main focus of research is to investigate the use of visual feedback for improving reaching and grasping capabilities of complex robots. To facilitate this a combined integration of computer vision and machine learning techniques is employed. From a robot vision point of view the combination of domain knowledge from both imaging processing and machine learning techniques, can expand the capabilities of robots. I present a novel framework called Cartesian Genetic Programming for Image Processing (CGP-IP). CGP-IP can be trained to detect objects in the incoming camera streams and successfully demonstrated on many different problem domains. The approach requires only a few training images (it was tested with 5 to 10 images per experiment) is fast, scalable and robust yet requires very small training sets. Additionally, it can generate human readable programs that can be further customized and tuned. While CGP-IP is a supervised-learning technique, I show an integration on the iCub, that allows for the autonomous learning of object detection and identification. Finally this dissertation includes two proof-of-concepts that integrate the motion and action sides. First, reactive reaching and grasping is shown. It allows the robot to avoid obstacles detected in the visual stream, while reaching for the intended target object. Furthermore the integration enables us to use the robot in non-static environments, i.e. the reaching is adapted on-the- fly from the visual feedback received, e.g. when an obstacle is moved into the trajectory. The second integration highlights the capabilities of these frameworks, by improving the visual detection by performing object manipulation actions.

Automated cross-modal mapping in robotic eye/hand systems using plastic radial basis function networks

Q. Meng and M. H Lee, Automated cross-modal mapping in robotic eye/hand systems using plastic radial basis function networks, Connection Science, 19(1), pp 25-52, 2007.

Biologically inspired automatic construction of cross-modal mapping in robotic eye/hand systems

Q. Meng and M.H. Lee, 'Biologically inspired automatic construction of cross-modal mapping in robotic eye/hand systems', IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS 2006,) ,4742-49, Beijing, 2006.

Psychologically Inspired Sensory-Motor Development in Early Robot Learning

M.H. Lee and Q. Meng, 'Psychologically Inspired Sensory-Motor Development in Early Robot Learning', in proceedings of Towards Autonomous Robotic Systems 2005 (TAROS-05), Nehmzow, U., Melhuish, C. and Witkowski, M. (Eds.), Imperial College London, 157-163, September 2005. See published version: http://hdl.handle.net/2160/485

Learning Commonsense Categorical Knowledge in a Thread Memory System

If we are to understand how we can build machines capable of broad purpose learning and reasoning, we must first aim to build systems that can represent, acquire, and reason about the kinds of commonsense knowledge that we humans have about the world. This endeavor suggests steps such as identifying the kinds of knowledge people commonly have about the world, constructing suitable knowledge representations, and exploring the mechanisms that people use to make judgments about the everyday world. In this work, I contribute to these goals by proposing an architecture for a system that can learn commonsense knowledge about the properties and behavior of objects in the world. The architecture described here augments previous machine learning systems in four ways: (1) it relies on a seven dimensional notion of context, built from information recently given to the system, to learn and reason about objects' properties; (2) it has multiple methods that it can use to reason about objects, so that when one method fails, it can fall back on others; (3) it illustrates the usefulness of reasoning about objects by thinking about their similarity to other, better known objects, and by inferring properties of objects from the categories that they belong to; and (4) it represents an attempt to build an autonomous learner and reasoner, that sets its own goals for learning about the world and deduces new facts by reflecting on its acquired knowledge. This thesis describes this architecture, as well as a first implementation, that can learn from sentences such as ``A blue bird flew to the tree'' and ``The small bird flew to the cage'' that birds can fly. One of the main contributions of this work lies in suggesting a further set of salient ideas about how we can build broader purpose commonsense artificial learners and reasoners.

Gradient-based reinforcement learning techniques for underwater robotics behavior learning

Darrerament, l'interès pel desenvolupament d'aplicacions amb robots submarins autònoms (AUV) ha crescut de forma considerable. Els AUVs són atractius gràcies al seu tamany i el fet que no necessiten un operador humà per pilotar-los. Tot i això, és impossible comparar, en termes d'eficiència i flexibilitat, l'habilitat d'un pilot humà amb les escasses capacitats operatives que ofereixen els AUVs actuals. L'utilització de AUVs per cobrir grans àrees implica resoldre problemes complexos, especialment si es desitja que el nostre robot reaccioni en temps real a canvis sobtats en les condicions de treball. Per aquestes raons, el desenvolupament de sistemes de control autònom amb l'objectiu de millorar aquestes capacitats ha esdevingut una prioritat. Aquesta tesi tracta sobre el problema de la presa de decisions utilizant AUVs. El treball presentat es centra en l'estudi, disseny i aplicació de comportaments per a AUVs utilitzant tècniques d'aprenentatge per reforç (RL). La contribució principal d'aquesta tesi consisteix en l'aplicació de diverses tècniques de RL per tal de millorar l'autonomia dels robots submarins, amb l'objectiu final de demostrar la viabilitat d'aquests algoritmes per aprendre tasques submarines autònomes en temps real. En RL, el robot intenta maximitzar un reforç escalar obtingut com a conseqüència de la seva interacció amb l'entorn. L'objectiu és trobar una política òptima que relaciona tots els estats possibles amb les accions a executar per a cada estat que maximitzen la suma de reforços totals. Així, aquesta tesi investiga principalment dues tipologies d'algoritmes basats en RL: mètodes basats en funcions de valor (VF) i mètodes basats en el gradient (PG). Els resultats experimentals finals mostren el robot submarí Ictineu en una tasca autònoma real de seguiment de cables submarins. Per portar-la a terme, s'ha dissenyat un algoritme anomenat mètode d'Actor i Crític (AC), fruit de la fusió de mètodes VF amb tècniques de PG.

Machine learning and small robot navigation

Developing successful navigation and mapping strategies is an essential part of autonomous robot research. However, hardware limitations often make for inaccurate systems. This project serves to investigate efficient alternatives to mapping an environment, by first creating a mobile robot, and then applying machine learning to the robot and controlling systems to increase the robustness of the robot system. My mapping system consists of a semi-autonomous robot drone in communication with a stationary Linux computer system. There are learning systems running on both the robot and the more powerful Linux system. The first stage of this project was devoted to designing and building an inexpensive robot. Utilizing my prior experience from independent studies in robotics, I designed a small mobile robot that was well suited for simple navigation and mapping research. When the major components of the robot base were designed, I began to implement my design. This involved physically constructing the base of the robot, as well as researching and acquiring components such as sensors. Implementing the more complex sensors became a time-consuming task, involving much research and assistance from a variety of sources. A concurrent stage of the project involved researching and experimenting with different types of machine learning systems. I finally settled on using neural networks as the machine learning system to incorporate into my project. Neural nets can be thought of as a structure of interconnected nodes, through which information filters. The type of neural net that I chose to use is a type that requires a known set of data that serves to train the net to produce the desired output. Neural nets are particularly well suited for use with robotic systems as they can handle cases that lie at the extreme edges of the training set, such as may be produced by "noisy" sensor data. Through experimenting with available neural net code, I became familiar with the code and its function, and modified it to be more generic and reusable for multiple applications of neural nets.

Incorporating proactivity to context-aware recommender systems for e-learning

Recommender systems in e-learning have proved to be powerful tools to find suitable educational material during the learning experience. But traditional user request-response patterns are still being used to generate these recommendations. By including contextual information derived from the use of ubiquitous learning environments, the possibility of incorporating proactivity to the recommendation process has arisen. In this paper we describe methods to push proactive recommendations to e-learning systems users when the situation is appropriate without being needed their explicit request. As a result, interesting learning objects can be recommended attending to the user?s needs in every situation. The impact of this proactive recommendations generated have been evaluated among teachers and scientists in a real e-learning social network called Virtual Science Hub related to the GLOBAL excursion European project. Outcomes indicate that the methods proposed are valid to generate such kind of recommendations in e-learning scenarios. The results also show that the users' perceived appropriateness of having proactive recommendations is high.

Low achieving mathematics students' attitudinal and achievement changes as a result of using an integrated learning system

This paper reports on a study to measure the effectiveness of an integrated learning system (ILS) in improving mathematics achievement for low achieving Year 5 to 9 students. The study found that statistically significant gains on the integrated learning system were not supported by scores on standardised mathematics achievement tests. It also found that although student attitudes to computers decreased (significantly for some items), the students still liked the integrated learning system and felt that it had helped them to learn.

How can we exploit mobile data to customise learning based on a student’s environmental surroundings?

User-Based intelligent systems are already commonplace in a student’s online digital life. Each time they browse, search, buy, join, comment, play, travel, upload, download, a system collects, analyses and processes data in an effort to customise content and further improve services. This panel session will explore how intelligent systems, particularly those that gather data from mobile devices, can offer new possibilities to assist in the delivery of customised, personal and engaging learning experiences. The value of intelligent systems for education lies in their ability to formulate authentic and complex learner profiles that bring together and systematically integrate a student’s personal world with a formal curriculum framework. As we well know, a mobile device can collect data relating to a student’s interests (gathered from search history, applications and communications), location, surroundings and proximity to others (GPS, Bluetooth). However, what has been less explored is the opportunity for a mobile device to map the movements and activities of a student from moment to moment and over time. This longitudinal data provides a holistic profile of a student, their state and surroundings. Analysing this data may allow us to identify patterns that reveal a student’s learning processes; when and where they work best and for how long. Through revealing a student’s state and surroundings outside of schools hour, this longitudinal data may also highlight opportunities to transform a student’s everyday world into an inventory for learning, punctuating their surroundings with learning recommendations. This would in turn lead to new ways to acknowledge and validate and foster informal learning, making it legitimate within a formal curriculum.

Scaffolding an EFL (English as a foreign language) ‘effective writing’ class in a hybrid learning community

This study sought to establish and develop innovative instructional procedures, in which scaffolding can be expanded and applied, in order to enhance learning of English as a Foreign Language (EFL) writing skills in an effective hybrid learning community (a combination of face-to-face and online modes of learning) at the university where the researcher is working. Many educational experts still believe that technology has not been harnessed to its potential to meet the new online characteristics and trends. There is also an urgency to reconsider the pedagogical perspectives involved in the utilisation of online learning systems in general and the social interactions within online courses in particular that have been neglected to date. An action research design, conducted in two cycles within a duration of four months, was utilised throughout this study. It was intended not only to achieve a paradigm shift from transmission-absorption to socio-constructivist teaching/learning methodologies but also to inform practice in these technology-rich environments. Five major findings emerged from the study. First, the scaffolding theory has been extended. Two new scaffolding types (i.e., quasi-transcendental scaffolding and transcendental scafolding), two scaffolding aspects (i.e., receptive and productive) and some scaffolding actions (e.g., providing a stimulus, awareness, reminder, or remedy) for EFL writing skills in an effective hybrid learning community have been identified and elaborated on. Second, the EFL ‘Effective Writing’ students used the scaffolds implemented in a hybrid environment to enhance and enrich their learning of writing of English essays. The online activities, conducted after the F2F sessions most of the time, gave students greater opportunities to both reinforce and expand the knowledge they had acquired in the F2F mode. Third, a variety of teaching techniques, different online tasks and discussion topics utilised in the two modes bolstered the students’ interests and engagement in their knowledge construction of how to compose English-language essays. Fourth, through the scaffolded activities, the students learned how to scaffold themselves and thus became independent learners in their future endeavours of constructing knowledge. Fifth, the scaffolding-to-scaffold activities provided the students with knowledge on how to effectively engage in transcendental scaffolding actions and facilitate the learning of English writing skills by less able peers within the learning community. Thus, the findings of this current study extended earlier understandings of scaffolding in an EFL hybrid learning environment and will contribute to the advancement of future ICT-mediated courses in terms of their scaffolding pedagogical aspects.

Learning the kernel matrix with semidefinite programming

Kernel-based learning algorithms work by embedding the data into a Euclidean space, and then searching for linear relations among the embedded data points. The embedding is performed implicitly, by specifying the inner products between each pair of points in the embedding space. This information is contained in the so-called kernel matrix, a symmetric and positive semidefinite matrix that encodes the relative positions of all points. Specifying this matrix amounts to specifying the geometry of the embedding space and inducing a notion of similarity in the input space - classical model selection problems in machine learning. In this paper we show how the kernel matrix can be learned from data via semidefinite programming (SDP) techniques. When applied to a kernel matrix associated with both training and test data this gives a powerful transductive algorithm -using the labeled part of the data one can learn an embedding also for the unlabeled part. The similarity between test points is inferred from training points and their labels. Importantly, these learning problems are convex, so we obtain a method for learning both the model class and the function without local minima. Furthermore, this approach leads directly to a convex method for learning the 2-norm soft margin parameter in support vector machines, solving an important open problem.

Open problems in the security of learning

Machine learning has become a valuable tool for detecting and preventing malicious activity. However, as more applications employ machine learning techniques in adversarial decision-making situations, increasingly powerful attacks become possible against machine learning systems. In this paper, we present three broad research directions towards the end of developing truly secure learning. First, we suggest that finding bounds on adversarial influence is important to understand the limits of what an attacker can and cannot do to a learning system. Second, we investigate the value of adversarial capabilities-the success of an attack depends largely on what types of information and influence the attacker has. Finally, we propose directions in technologies for secure learning and suggest lines of investigation into secure techniques for learning in adversarial environments. We intend this paper to foster discussion about the security of machine learning, and we believe that the research directions we propose represent the most important directions to pursue in the quest for secure learning.

Learning from episodes of degradation and recovery in variable Australian rangelands

Land-change science emphasizes the intimate linkages between the human and environmental components of land management systems. Recent theoretical developments in drylands identify a small set of key principles that can guide the understanding of these linkages. Using these principles, a detailed study of seven major degradation episodes over the past century in Australian grazed rangelands was reanalyzed to show a common set of events: (i) good climatic and economic conditions for a period, leading to local and regional social responses of increasing stocking rates, setting the preconditions for rapid environmental collapse, followed by (ii) a major drought coupled with a fall in the market making destocking financially unattractive, further exacerbating the pressure on the environment; then (iii) permanent or temporary declines in grazing productivity, depending on follow-up seasons coupled again with market and social conditions. The analysis supports recent theoretical developments but shows that the establishment of environmental knowledge that is strictly local may be insufficient on its own for sustainable management. Learning systems based in a wider community are needed that combine local knowledge, formal research, and institutional support. It also illustrates how natural variability in the state of both ecological and social systems can interact to precipitate nonequilibrial change in each other, so that planning cannot be based only on average conditions. Indeed, it is this variability in both environment and social subsystems that hinders the local learning required to prevent collapse.