Dragana Laketic

Stochastic Adaptation to Environmental Changes Supported by Endocrine System Principles

Adaptation to a changing environment presents a challenging task for todays electronic systems when operating in dynamic, fluctuating environments. This applies in particular to the systems which are to operate in harsh environments with no possibility for human intervention when the change occurs. As a rule, environmental changes are stochastic and so is the effect they may exhibit on a man-made system. On the other hand, evolution has provided living organisms with in-built mechanisms for adapting to environmental changes. In particular, homeostatic processes are example of such inherent adaptive mechanisms found within human body. Out of many complex and interweaved systems involved in homeostatic processes, hormones and endocrine system are prominent for certain properties they exhibit. In the first place, this refers to communication within the system and control of the regulatory processes, both of which are challenging issues within man-made systems. This paper investigates endocrine system principles applied within adaptive processes in a man-made system when adaptation is of stochastic nature. Presented results refer to applications in systems of modular architecture.

Adaptation in tissue sustained by hormonal loops

When faced with the requirements for the design of an autonomous adaptive system, many aspects of the system organisation need be addressed. In living systems, the co-evolution with the environment has provided the solution for such challenges in a form of inherent mechanisms which are employed when the environmental fluctuation occurs leading to the organism achieving adaptation through some adaptive process. In this paper we investigate such mechanisms, more precisely the priniples on which their operation is based. In particular, the focus is set on endocrine system within homeostatic processes. We postulate that adaptation to a fluctuating environment can be achieved if initiated and sustained by the hormone flow loops. Such statement is further supported by simulations. Based on the recognised advantages of the system organisation endowed with the ability to form hormonal loops, the avenues of research are identified for further work.

Extreme Temperature Electronics - from Materials to Bio-inspired Adaptation

Biological systems have inherent mechanisms which ensure their adaptation and thus survival - preservation of functionality, despite extreme and varying environments. One such environmental feature is that of temperature. Extreme temperature electronics (ETE) is afield where, similarly, these organisms (electronic solutions), have to be designed to survive in such an environment. A number of approaches that address ETE are both proposed and, in some cases, implemented in todays technologies. Some of these approaches may be said to reduce this challenge but none may be said to solve it. However, biology has found a solution. There can, therefore, be great merit in turning to biology to identify possible solutions. However, it is important to first consider where the field is today. This paper presents a survey of methods and techniques for tackling temperature effects in ETE - from materials to static and dynamic design techniques. Further, suggestions are provided as to where a bio-inspired approach may be applied either as an improvement to an existing approach or as a novel approach to an existing sub-challenge. Particular attention has been given to where a bio-inspired approach might provide a more dynamic solution.

Autonomous adaptation inspired by the model of a minimal living system provided by chemoton theory

Abstract Future computing machines will have to meet increasing requirements regarding the computational power and the efficient use of resources. Whatever the technology may be, in all likelihood it will be based on parallel operation of a large number of interconnected nanoscale units. Further challenges lie in the choice of basic units and their mutual communication. Moreover, an additional design challenge comes from the sensitivity to environmental variations which is pronounced at such a low scale. Biological creations are living examples of similar designs—they are built of a number of cells, numbers ranging from one to thousands of millions. The cells are organised in a particular way and interconnected by subtle mechanisms in achieving the ultimate common goal—the preservation of viability. In doing so, living systems incessantly adapt to ever-varying environments. In this paper, we investigate adaptive mechanisms at a very low level–the protocell level–and consider a minimal living system in a form provided by chemoton theory by Tibor Ganti. We suggest that adaptive traits of the Chemoton be used as guidelines for the design of an adaptive cell within a modular man-made system. As a proof of concept, we propose a basic circuitry in silicon and argue in favour of such implementation of the proposed adaptive cell.

Evolution-in-materio computations: Hierarchies rising from electron dynamics in carbon nanotubes

Previously, Evolution-In-Materio (EIM), an unconventional computing paradigm, was addressed as a computing system which exhibits dynamical hierarchies. For different conceptual domains identified within an EIM system, a corresponding hierarchical level was defined and the state space description provided. Entropic relations established between such system descriptions show that one hyperdescribes another in an information theoretical sense. Hereby we report on those findings and revisit entropic relations between the level of material physics and the level of measurements via simulations of the addressed physical phenomenon.

A hierarchical view on Evolution-In-Materio computations

Evolution-In-Materio, an unconventional computing paradigm exploiting physical properties of materials for achieving computations, is addressed here as a system which exhibits dynamical hierarchies. A description of computations is provided to show that computations within Evolution-In-Materio systems arise from the dynamics at different hierarchical levels. An information theoretic approach to formalising the notion of dynamical hierarchies is used. The approach is based on the descriptions of the system at different hierarchical levels. The concrete material addressed in this paper is a carbon nanotube / polymer nanocomposite. The choice of material is based on previous work motivated by a number of experiments conducted on such material samples. Presented findings are valuable for several reasons: better understanding of computations within Evolution-In-Materio systems, useful hints to modelling this kind of unconventional computations, useful ideas for further development of similar unconventional computing systems such as using quantum properties of charge carriers within the material and the magnetic field for guiding the search for the solution of the computational problem at hand.