ABSTRACT. Vector Symbolic Architectures VSA is an approach for encoding and operations on distributed representation of knowledge, and has previously been used mainly in the area of cognitive computing and natural language processing. The fundamental difference between the distributed and traditional (localist) representations of data is as follows: in localist computing architectures each bit and its position within a structure of bits are significant (for example, a field in a database has a predefined offset amongst other fields and a symbolic value has unique representation in ASCII codes); whereas in a distributed representation all entities are represented by vectors of very high dimension. That is, entities are represented by the direction of a vector in a high-dimensional space and every bit contributes to defining the direction of the vector. In a localist representation single bits or small groups of bits can be interpreted without reference to the remainder of the bits. In a distributed representation it is only the total set of bits that can be interpreted. High dimensionality refers to that fact that several thousand positions (of binary numbers) are used for representing a single entity. In this talk the essentials hyper-dimensional computing and vector symbolic architectures, their properties and selected illustrative applications will be presented.

ABSTRACT. Nearly all experimental results in modern science are the results of indirect measurements. This means that there is a separate problem called an inverse problem (IP) - to extract the information interesting for the researcher from the measured experimental data. Solving such problems is an inherent necessity in many areas of science, including spectroscopy, geological prospecting, aerospace image processing and others. The lecture discusses methodological aspects of the solution of IP with the help of such biologically inspired cognitive architecture as artificial neural networks (ANN). Different formulations of IP from the point of view of data processing methods are given. Various methodological approaches to the solution of IP using ANN techniques called “experiment-based”, “model-based”, and “quasi-model” approaches, are considered. Their characteristics, differences and areas of application are discussed. The differences of ANN from other methods of solution of IP are discussed, as well as the key areas where their use is justified. Different approaches to simultaneous determination of parameters when solving multi-parameter IP are considered. The material is illustrated by examples of IP from the areas of optical spectroscopy and electrical prospecting.

ABSTRACT. The cognitive architecture designed within the Natural-Constructive Approach (NCA) to modeling the cognitive process is presented. This approach is based on the dynamical theory of information, the neurophysiology data, and neural computing (combined with nonlinear differential equation technique for the concept of dynamical formal neuron). Main peculiar feature of the architecture is split of the whole system into two subsystems, by analogy with two cerebral hemispheres of human brain. One of them should necessarily contain the random element (noise) required for generation of information; being responsible for creative work and learning. The other one, being free of noise, is responsible for processing the well-known information. It is shown that this architecture provides the possibility to interpret and reproduce peculiar features of the human cognitive process, namely – uncertainty, individuality, intuitive and logical thinking, etc. Within NCA, the emotions could be associated with the noise-amplitude variation, and this very variation does control the activity of two subsystems (i.e., the cross-subsystem connections corresponding to corpus callosum). The neurophysiology-inspired model for coupling up the noise amplitude with the specific variable (corresponding to effective composition of neurotransmitters) is imbedded into the architecture. It is shown that this model provides the possibility to describe, in particular, the effect of stress/shock under extreme external conditions, which is in qualitative agreement with experiment.

ABSTRACT. In 1998, Philips coined the term «Ambient Intelligence» (AmI) in order to illustrate a vision of the future where various information technologies seamlessly interact and adapt to human needs while being none obtrusive. Ambient Intelligence systems aim at augmenting real world environments to create Smart Spaces where users are provided with pervasive virtual services. According to classical definition, an AmI system is a digital environment that proactively, but sensibly, supports people in their daily lives. Nowadays, Ambient Intelligence is often used in a wider sense as a generic concept, and the close term «Smart Environment» is reserved to describe the physical infrastructure (e.g. sensors, actuators and networks) that supports the AmI-system. It is worth noticing that AmI represents a first step towards the implementation of NBICS-convergence technologies (NBICS are the first letters to denote the following technologies: N – Nano, B – Bio, I – Info, C – Cogno, S –Socio). The developments in these fields not just complement each other – the fields are gradually merging. Such a fusion of these branches corresponds to the concept of synergistic science where heterogeneous disciplines and technologies co-operate to enable a new seamless user-friendly artificial environment of smart networked devices. So AmI scientific problems are faced on the crossroad of computer science, cybernetics, ergonomics, artificial intelligence, cognitive sciences, behavior sciences, social sciences, mechatronics and robotics. Primarily, a cybernetic problem of organizing the required behavior of artificial micro-environment is solved by using both negative feedback and positive feedback; here such problems as user recognition, needs awareness, behavior context understanding, as well as the comparison of real control results with expected ones, are of special concern. Also the black-box concept is conserved augmented by Weiser’s ideas: “The most profound technologies are those that disappear. They weave themselves into the fabric of everyday life until they are indistinguishable from it”. Thus, the miniaturization of embedded sensors and actuators brings about their dissolution in real environment (or disappearance), so the people perceive only the friendly user interface. Besides, we deal with a new generation of «man-machine-environment» systems usually studied in ergonomics. Indeed, the AmI represents a new concept of man-machine interface, where people are surrounded by useful micro-devices embedded into physical world and tied to intelligent software. Thus, a hybrid intelligent cyber-physical environment emerges based on invisible collective (web) intelligence. The AmI technological devices ought to be embedded, ubiquitous, context aware, personalized, transparent, anticipatory, well-adapted to human senses Generally, technological resources of AmI are related to the following six areas: 1) sensors and pervasive measurement devices; 2) actuators, mechatronic devices and robotic systems; 3) ubiquitous networks; 4) ubiquitous computing; 5) user friendly (anthropocentric) interface; 6) Artificial Intelligence (AI) and Multi-Agent Systems. A variety of AmI concepts, models and architectures is presented in our lecture. Specifically, some biologically and psychologically inspired AmI architectures are considered.

ABSTRACT. In the Lecture on the examples of specific problems the principles of neuro control in a robotics are considered. Tasks which are discussed: the motion of the robot along the colored line, problems of navigation (localization) for biped and the six-legged walking robots, a problem of control of a quadrocopter in the modes take-off-soaring-landing, flight along trajectories, aerobatics. The last tasks represent an important class of problems of nonlinear control of systems with deficiency of control directors of influences (nonlinear control of underactuated systems). At the beginning of lecture short introduction to technology of neural networks is given, types of networks and their opportunity and advantages to robotic control are discussed.

ABSTRACT. The lecture describes current models of autonomous cognitive agents. The study of these models can be considered as the method of investigations of biologically inspired cognitive architectures (BICA). The main attention is paid to the models that are used at studying of cognitive evolution. Several examples of such models are outlined. Schemes of new models are proposed.

ABSTRACT. The Intelligent System that implements the plausible reasoning simulates and intensifies productive thinking to discover new knowledge.