Georgi Stojanov

Interactionist-expectative view on agency and learning

In this paper, we present a theory of agency with interaction and expectation as its central concepts. The research was motivated by our work in cognitive brain waves and expectograms, and heavily influenced by the works of Tolman and Piaget. Our intention is to provide remedies for the observed deficiencies in the existing approaches to the problem of autonomous agents (AA) design. A brief critical overview of the relevant agent oriented research is given first. It is followed by the presentation of the theory and an instantiation of it through a simulated AA called “Petitage” (Pee-ah-zhe). Algebraic framework is presented as a basis for the simulation experiments. The final part of the paper is devoted to the results and discussion of the simulation experiments.

Embodiment as metaphor: metaphorizing- in the environment

The paper describes a general mechanism for internalization of environment in autonomous agents. After reviewing the role of representation in behavior-based autonomous agents, we propose metaphor framework that unifies various research threads in the domain. We start from a variant of the so-called similarity creating metaphors for the case of implicit target domain (the environment of the agent). The mechanism is based on a fairly simple idea of assimilation via inborn schemas as understood in Piagets developmental psychology. These schemas represent the source domain for the metaphor. They are ordered sequences of elementary actions that the agent is capable of performing. Because of the environmental constraints, when the agent tries to execute some schema, only certain subsequences from the original schema will actually be performed. These subsequences are called enabled schema instances. Thus, environment unfolds its structure to the agent via the subset of the enabled schema instances. Another way to look at this is to say that what the agent gets is a metaphorical description of its environmental niche (the implicit target domain) in terms of instances of its inborn schemas (the source domain). After describing the basic idea by means of an example, we present some simulation results that show the plausibility of the model. The simulated agent solves the navigational problem in an initially unknown environment. The paper closes with a discussion section where we compare our model with some related works and make the case for the metaphor framework as a proper unifier of diverse research work in embodied and situated cognition.

Inductive Logic Programming ILP and Reasoning by Analogy in Context of Embodied Robot Learning

The ability to reason by analogy is essential for many cognitive processes from low-level and high-level perception to categorization. Intuitively, the idea is to use what is already known to explain new observations that appear similar to old knowledge. In a sense, it is opposite of induction, where to explain the observations one comes up with new hypotheses/theories. Therefore, a system capable of both types of reasoning would be superior. In this paper, the authors present an overview of Inductive Logic Programming ILP systems that use reasoning by analogy and discuss the results of combining Analogical Prediction with an ILP system, showing that, for some cases, it is possible to improve significantly the learning speed of the ILP system. This paper will examine the problems that arise in the context of a physically embodied robot that tries to learn regularities in its environment.

Computer Interfaces: From Communication to Mind-Prosthesis Metaphor

This paper explores underlying metaphors about the process of working with computers and computer interfaces. First, we identify the most wide-spread implicit metaphor, which we claim to be the conduit metaphor. While interacting with a computer, users are implicitly put in a conversational situation, unlike the situation they find themselves in when interacting with most other artifacts. We advance arguments for this thesis taken from the history of computers, computer interfaces, and their current design. Nowadays we are witnessing a shift from this, all pervasive metaphor towards another emerging metaphor where computers are beginning gradually to be perceived as an augmantation, or prosthesis for the perceptive and cognitive capabilities. In this, transition phase, we can see people advocating views where the two metaphors are mixed. Then, we put forward the claim that the prosthesis metaphor is far more fruitful, productive, and explicative and we indicate some of the practical implications of adopting this metaphor.

The Higher Education System in Macedonia: Overview, Reforms, and Prospects

The system of higher education in Macedonia is undergoing the largest reform process since the mid-1940s. The context is rather turbulent. Fourteen years after Macedonia’s separation from the Yugoslav Federation, this small country now deals with an unstable political situation (the Kosovo question) and awaits an answer from Brussels (expected in November 2005) as to whether it will be accepted as a candidate country to join the European Union [EU]. Macedonia is burdened with a heavy economic recession (unemployment approaches a staggering 40 percent), and is attempting to apply the Bologna Declaration recommendations to restructuring its higher education system, as it became one of the forty-three signatory countries in September 2003. All these elements are interdependent, as will be seen below.

Learning patterns in wireless sensor networks based on wavelet neural networks

In this paper it is demonstrated how some of the algorithms developed within the artificial neural-networks tradition can be simply adopted to wireless sensor network platforms and still meet most of the requirements for sensor networks. Neural-networks clustering algorithms also provide dimensionality reduction which further leads to lower communication costs and thus bigger energy savings. Two different data aggregation architectures are presented. They both utilize algorithms which apply wavelets for initial data-processing of the sensory inputs at different resolutions. Artificial neural-networks which make use of unsupervised learning methods are used for categorization of the sensory inputs. These architectures are tested on a data obtained from a set of several motes, equipped with several sensors each. Results from simulations of intentionally made defective sensors demonstrate the data robustness of these architectures

Perceptual Similarity and Analogy in Creativity and Cognitive Development

We argue for the position that analogy represents the core mechanism in human cognitive development rather than being a special cognitive skill among many. We review some developmental psychology results that support this claim. Analogy and metaphor, on the other hand, are seen as central for the creative process. Whereas mainstream research in artificial creativity and computational models of reasoning by analogy stresses the importance of matching the structure between the source and the target domains, we suggest that perceptual similarities play a much more important role. We provide some empirical data to support these claims and discuss their consequences.

A model of an expectancy-driven and analogy-making actor

A model of autonomous actors is presented as an attempt to build a more encompassing cognitive architecture. Various cognitive phenomena, like expectations, planning, curiosity, analogy-making, motivation and attention, are encountered by the model. The actors knowledge about the environment is represented in a pseudo-conceptual network whose parts can be activated at different levels of activation. The nodes of this network are formed on the basis of the sensory inputs, while the links are formed on the basis of the motor actions. Generally, the actor has two main subsystems, a contemplating and a behavioral subsystem. The contemplating subsystem is responsible for problem solving by means of analogy-making, while having little or no influence from the environment. The behavioral subsystem is responsible for the execution of the pre-selected behavioral plans.

Inductive Logic Programming and Embodied Agents: Possibilities and Limitations

Open-ended learning is regarded as the ultimate milestone, especially in intelligent robotics. Preferably it should be unsupervised and it is by its nature inductive. In this article we want to give an overview of attempts to use Inductive Logic Programming ILP as a machine learning technique in the context of embodied autonomous agents. Relatively few such attempts exist altogether and the main goal in reviewing several of them was to find a thorough understanding of the difficulties that the application of ILP has in general and especially in this area. The second goal was to review any possible directions for overcoming these obstacles standing on the way of more widespread use of ILP in this context of embodied autonomous agents. Whilst the most serious problems, the mismatch between ILP and the large datasets encountered with embodied autonomous agents seem difficult to overcome we also found interesting research actively pursuing to alleviate these problems.

Sound and video processing in wireless sensor networks

The general problem of data management in wireless sensor networks (WSNs) is to provide efficient aggregation of different sensors data taking into account the problems of the limited energy of the nodes and their unpredictable failures. Generally, this is solved by reducing the communication among nodes. In order to have an efficient data aggregation performance, a pre-processing is needed which would reduce the amount of data being sent over the communication channels. As an outcome of this research, we propose two similar architectures for data aggregation of sound and video signals. These classification architectures have the same core consisted of a modified FuzzyART neural network and a modified SEQUITUR algorithm used previously only for analysis of symbolic sequences. The proposed architectures have been tested in a prototype implementation using Pocket PCs having microphones and cameras as sensors