Publications related to intelligent devices.
- E. I. Barakova, J. Gillessen, and L. Feijs Social training of autistic children with interactive intelligent agents [877 KB pdf]. Journal of Integrative Neuroscience, 8(1):23-34, 2009.
Abstract
The ability of autistic children to learn by applying logical rules has been used widely in behavioral therapies for social training. We propose to teach social skills to autistic children through games that simultaneously stimulate social behavior and include recognition of elements of social interaction. For this purpose we created a multi-agent platform of interactive blocks, and we created appropriate games that require shared activities leading to a common goal. The games included perceiving and understanding elements of social behavior that non-autistic children can recognize. We argue that the importance of elements of social interaction such as perceiving interaction behaviors and assigning metaphoric meanings has been overlooked, and that they are very important in the social training of autistic children. Two games were compared by testing them with users. The first game focused only on the interaction between the agents and the other combined interaction between the agents and metaphoric meanings that are assigned to them. The results show that most of the children recognized the patterns of interaction as well as the metaphors when they were demonstrated through embodied agents and were included within games having features that engage the interest of this user group. The results also show the potential of the platform and the games to influence the social behavior of the children positively.
- E. I. Barakova Emotion recognition in robots in a social game for autistic children [113 KB pdf]. In J. Sturm and M.M. Bekker, editors, Proceedings of the 1st workshop on Design for Social Interaction through Physical Play, pages 21-25. Eindhoven, the Netherlands, 2008.
Abstract
This paper provides a framework for a social game that has as a goal improving the social interaction skills through associative play. It describes the design of the game platform and an ongoing study on the perception of emotional expression from motion cues for communication and social coordination. Especially, children with autism spectrum disorders are targeted, since they will benefit most from behavioral training that may improve their social skills. The promising results from two stages of this work are shown.
- E. I. Barakova, G. van Wanrooij, R. van Limpt, and M. Menting Using an emergent system concept in designing interactive games for autistic children [401 KB pdf]. 6th International Conference on Interaction Desing and Children (IDC07), pages 73-77, Aalborg Denmark, June 2007. ACM 978-1-59593-747-6.
Abstract
This paper features the design process, the outcome, and preliminary tests of an interactive toy that expresses emergent behavior and can be used for behavioral training of autistic children, as well as for an engaging toy for every child. We exploit the interest of the autistic children in regular patterns and order to stimulate their motivational, explorative and social skills. As a result we have developed a toy that consists of undefined number of cubes that express emergent behavior by communicating with each other and changing their colors as a result of how they have been positioned by the players. The user tests have shown increased time of engagement of the children with the toy in comparison with their usual play routines, pronounced explorative behavior and encouraging results with improvement of turn taking interaction.
- J. Nijhuis, H. van Aartsen, E. I. Barakova, W. Jansen, and B. Spaanenburg On the optimal Mapping of Fuzzy Rules on standard Micro-Controllers [327 KB pdf]. Microprocessing and microprogramming 40:697-700, 1994.
Abstract
Once a fuzzy controller is specified by a rule-set, it can be implemented in dedicated hardware or as a software program. For industrial applications, an inexpensive micro-controller with limited resources is often selected. The implementation (or mapping) issue then leads to a trade off between operation speed and memory usage. This paper presents a set of basic transformation rules that allows the designer to optimize such a mapping.
- E. I. Barakova, J. C. C. Gillesen, and L. M. G. Feijs Use of goals and dramatic elements in behavioral training of children with ASD [303 KB pdf]. Proceedings of the 7th international conference on Interaction design and children, pages 37-40, 2008. ACM, New York.
Abstract
We describe the development of a multi-agent platform and adequate games that aim to stimulate social behaviro of autistic children. User tests with two games, one with emerging patterns and another with goals and dramtaic elements were compared. The results show that the childeren do not play significantly longer with either of the games, when exposed for first time to the multi-agent toy. Interestingly, most of the children recognized the dramatic elements, which makes us believe that by longer exposure and proper guidance children might be thought social skills. Test results are described quantitatively and qualitatively.
- B. van der Vlist, R. van de Westelaken, C. Bartneck, J. Hu, R. Ahn, E. I. Barakova, F. Delbressine, and L. Feijs Machine Learning to Design Students [150 KB pdf]. Technologies for E-Learning and Digital Entertainment, pages 206-217, 2008.
Abstract
Machine learning is a key technology to design and create intelligent systems, products, and related services. Like many other design departments, we are faced with the challenge to teach machine learning to design students, who often do not have an inherent affinity towards technology. We successfully used the Embodied Intelligence method to teach machine learning to our students. By embodying the learning system into the Lego Mindstorm NXT platform we provide the student with a tangible tool to understand and interact with a learning system. The resulting behavior of the tangible machines in combination with the positive associations with the Lego system motivated all the students. The students with less technology affinity successfully completed the course, while the students with more technology affinity excelled towards solving advanced problems. We believe that our experiences may inform and guide other teachers that intend to teach machine learning, or other computer science related topics, to design students.
- L. Feijs and E. I. Barakova Semantics through Embodiment: a Non-linear Dynamics approach to Affective Design [1577 KB pdf]. In L. Feijs, S. Kyffin, and B. Yong, editors, DesForm 2007, pages 108-116, 2007.
Abstract
In this paper we address the creation and interpretation of movements, light and sound from a fundamental and innovative viewpoint. Using a number of concepts from the relatively new and very promising research field of nonlinear adaptive systems, and getting some inspiration from psychophysical studies on the perception of emotion we address the study of movements and other autonomous expressions of products. The goal is to understand the semantics of movement, particularly the emotional meaning of the movement and to translate it to other autonomous expressive behavior.
- E. I. Barakova Emergent behaviors based on episodic encoding and familiarity driven retrieval [431 KB pdf]. In C. Bussler and D. Fensel, editors, Artificial Intelligence: Methodology, Systems, and Applications, 11th International Conference, AIMSA 2004, volume 3192 of Lecture Notes in Artificial Intelligence, pages 188-197. Springer-Verlag, 2004.
Abstract
In analogy to animal research, where behavioral and internal neural dynamics are simultaneously analysed, this paper suggests a method for emergent behaviors arising in interaction with the underlying neural mechanism. This way an attempt to go beyond the indeterministic nature of the emergent behaviors of robots is made. The neural dynamics is represented as an interaction of memories of experienced episodes, the current environmental input and the feedback of previous motor actions. The emergent properties can be observed in a two staged process: exploratory (latent) learning and goal oriented learning. Correspondingly, the learning is dominated to a different extent by two factors: novelty and reward. While the reward learning is used to show the relevance of the method, the novelty/familiarity is a basis for forming the emergent properties. The method is strongly inspired by the state of the art understanding of the hippocampal functioning and especially its role in novelty detection and episodic memory formation in relation to spatial context.
- E. I. Barakova Learning Reliability: a study on indecisiveness in sample selection [3687 KB pdf]. PrintPartners Ipskamp B.V. ISBN 90-367-0987-3, March 1999.
Preface
The design of a product is based on the assumption of how it will be used. Conversely, the product is usually good for only such usage as was assumed during its conception. In a classical sense, the implicit assumption brings an explicit specification from which the design is derived. More often than not, the specification is therefore the starting point of a hopefully structured and well–behaved, but eventually mechanical design effort. Where the customer tends to learn from the design and mandates to change and/or augment the specification during the process, the project planning gets invalidated. Current practice is therefore to fix the specification in advance, for instance by contract.
The interest in Artificial Neural Networks (ANN) is founded on their ability to learn from examples, as derived from the environment in which the product will operate, instead of being designed from an hypothesis about the operation. It is commonly agreed that learning is based on memorization (associating or mapping a set of questions to their answers) and generalization (the ability to answer new questions about the same problem). As such, ANNs promise a perfect fit to their intended usage. But circumstantial evidence still does not equal a witness observation. Despite its historic fame, an Artificial Neural Network will not learn all, let alone under all circumstances. This is probably the most striking difference with a designed product: there will never be a proof by construction!
With the coming of age of neural technology, an impressive number of neural products have found their way to the market place [88]. Some popular applications are indicated in figure 1, which position them in the area spanned by computational complexity and model correctness. The bars indicate the achieved performance: the patterns within the bar indicate the widely achieved results, since the white part stands for the best results in the area. Clearly, none of them achieves a 100% correct functionality. It appears, that for each application a bottom level of functionality can be reached almost without any effort. However, to go beyond requires special attention and has therefore spurred a lot of research to develop new algorithms, to construct alternative architectures, to provide different settings of input parameters or to preprocess input data.
To achieve a product of ultimate performance, two methods can be devised: (a) its function is based on a provably correct algorithm, and (b) an effective redundancy is to be incorporated in the underlying algorithm. As far as ANNs are constructed from analysis of noisy data, they can entirely be considered as systems of the second type. Because statistics is concerned with data analysis as well, there is a considerable overlap between the fields of neural networks and statistics. To analyze learning and generalization of neural networks from noisy/randomized data, statistical inference can also be used.
Performance enhancement can be created by a kind of majority voting. This principle suggests that, instead of providing one neural network solution to a problem, a set of neural networks can be combined to form a neural net system which performs better than any of the networks on its own [116] [138]. The conclusion made in [112] is that mere redundancy does not necessarily increase reliability. Empirically it is common practice to train many different candidate networks to select the winner on basis of predefined criteria. A disadvantage of this method is that training of the losing networks does not help in a further development. Another weak point is that the criterion for choosing the best network is usually the performance on a validation set, which can not guarantee the modeling quality of the underlying data generator. But when the networks are incomplete versions of the same functionality, the combination might raise the functional correctness to a higher level (Figure 2).
The committee arrangement generalizes this idea. It can have significantly better predictions on new data at an acceptable increase of the computational complexity. The performance of the committee can be much better than the performance of each single network in isolation. The committee contains a set of a trained networks diversified in a distinct way. Diversity can appear in the number of hidden neurons, in the kind of network model, in the mixture of networks, in the optimization criteria, in the initial weight configuration, training parameters in the training samples, etc. The extent to which reliability can be improved by combining neural net solutions depends on the type of diversity, present in the set of nets.
All such techniques assume that the basic neural network is optimally trained. However, we have noticed that training algorithms are often slow and sometimes unable to converge, even though the underlying techniques often perform very well on other problems. In other words, even though an ANN can be trained to some functionality, there appears to be an underlying problem that causes unreliability in learning. This thesis will therefore be devoted to unravel such circumstances and to contribute ways in which reliable learning can be achieved. By large, the neural paradigm problem is represented as a stream of examples (data) and that guides the learning algorithm to adapt the network parameters until the network is trained to give the right answers to the posed questions. Thus the success and the reliability of this training depends to a large extent on the content and composition of this data stream.
Overall unreliable learning can be considered to result from the interaction between three factors: network, problem, and algorithm. In an attempt to answer questions like why and when the learning process will become unreliable and when a systematic failure can appear, backpropagation (still the algorithm with highest practical significance) has been used. The restricted class of architectures it is supposed to be used for and the feedforward architecture allow us to elaborate in more detail on the problem with respect to the chosen architecture and algorithm.
As we found that the conventional focus on network, problem and algorithm leaves much to be desired, we propose here to base the discussion rather on symmetry, randomness (as basic network design principles), and knowledge (the problem to be learned) as the basic ingredients of the universe of discourse. A high degree of symmetry in the initially designed network is historically viewed to favor the learning algorithm in providing an equal chance to move in several directions. However, this has also a drawback: the freedom of choice may lead to indecisiveness. Admittedly, randomness may in turn help the network escape from such a dilemma. But then again, randomness may wipe away the knowledge; hence a working balance should be found.
Symmetry can be dominant in the beginning of, but also at specific moments during, learning. Randomness (for instance as stochastic variable in the learning algorithm or as additional noise at the network input, output or internal parameters ) is then required to force the presentation of examples to follow alternative itineraries. When the amount of randomness is not sufficient to counteract symmetry, learning will not be completed: instead of being adapted to ensure the right mapping between input/output data strings, the initial parameters will eventually become zero. If the noise (the randomness) of the system is dominant, learning will also be unsuccessful, because the network will rather learn the noise than the exemplified knowledge. The fundamental issue of learning is therefore the creation of a functional balance between symmetry and randomness directed by the examples (the knowledge).
To bring this idea into tangible borders, the interaction between learning components is represented in the error surface paradigm. The network will be able to extract the necessary information by adapting itself to map the questions posed to the right answers. This adaptation is in fact an optimization procedure and is thus equivalent to finding the minimum energy state on an error landscape. The steps, that the learning algorithm takes on this landscape, are directed by the presented examples and form a learning trajectory on this surface. Directing this itinerary properly can help to escape some difficulties to pass surface areas, at which the learning algorithm normally spends a lot of time on or from which it can never escape. For finding an optimal trajectory on the error surface, the so–called regularisation methods have been used. An alternative effect has the introduction of extra noise during training. Our objection here is that the task complexity or the convergence accuracy may be changed in an unwanted direction. The investigation of the statistical long–run effects of example presentation when traveling on the difficult forms of the global error surface brings us to a constructive algorithm which helps in escaping them.
Therefore, the work in this thesis takes an alternative route to ensure reliable learning by focussing on sample diversity [116]. On basis of the instantaneous characteristics of the current training set we will conclude on learnability, reorder the set if necessary to establish the best sample sequence and train eventually a single network with success.
In conclusion, this thesis aims to give directions on how learning can be guaranteed so that its duration will be short and stable and its success unquestionable from the outset. In this respect, we aim to contribute to move neural technology from the realm of ‘Learning by Examples’ to ‘Design by Examples’. - S. H. M. Alers, E. I. Barakova Multi-agent platform for development of educational games for children with autism [625 KB pdf]. IEEE ICE CIG 2009, in press, 2009.
Abstract
Multi-agent system of autonomous interactive blocks that can display its active state through color and light intensity has been developed. Depending on the individual rules, these autonomous blocks could express emergent behaviors which are a basis for various educational games. The multi-agent system is used for developing games for behavioral training of autistic children. This paper features the functional and electronic design of the individual blocks and transformation of the multi-agent system to a platform that allows multiple games to be designed through easy reprogramming of the blocks. Three game concepts that show the type of games that can easily be implemented and reprogrammed are described. The impact of this platform is shortly mentioned in the discussion. The initial tests of using the platform for various educational games are very positive. However, the results of user tests go beyond the scope of this paper and are not discussed in the text that follows.