Skip to main content

Über dieses Buch

Products of modern artificial intelligence (AI) have mostly been formed by the views, opinions and goals of the “insiders”, i.e. people usually with engineering background who are driven by the force that can be metaphorically described as the pursuit of the craft of Hephaestus. However, since the present-day technology allows for tighter and tighter mergence of the “natural” everyday human life with machines of immense complexity, the responsible reaction of the scientific community should be based on cautious reflection of what really lies beyond AI, i.e. on the frontiers where the tumultuous ever-growing and ever-changing cloud of AI touches the rest of the world.

The chapters of this boo are based on the selected subset of the presentations that were delivered by their respective authors at the conference “Beyond AI: Interdisciplinary Aspects of Artificial Intelligence” held in Pilsen in December 2011.

From its very definition, the reflection of the phenomena that lie beyond AI must be inherently interdisciplinary. And so is this book: all the authors took part in a mutual transdisciplinary dialogue after explaining their views on AI not only to a narrow selection of their usual close peers with the same specialisation, but to a much broader audience of various experts from AI engineering, natural sciences, humanities and philosophy. The chapters of this book thus reflect results of such a dialogue.



Future of Artificial Intelligence


On the Way to Intelligence Singularity

Since the fifties of the last century there have been debates about the so called “technological singularity”, motivated by the predicted and later actual exponential growth of the speed and power of computers. Recently the interest of futurologists and philosophers shifts to the so called ‘intelligence singularity’ which some of them predict to happen soon after human intelligence is surpassed by artificial intelligence. This study critically analyzes certain assumptions behind the concept of intelligence singularity, in particular the idea of explosive growth of intelligence of machines with the ability of designing machines more intelligent than themselves.
Ivan M. Havel

Slippage in Cognition, Perception, and Action: From Aesthetics to Artificial Intelligence

A growing body of work has emerged in computer science and related fields around the topics of aesthetics and affect. Much of this work has focused on the issue of how to treat computational systems aesthetically rather than on the question of how to understand aesthetics computationally. Here we pursue the latter question, exploring it through the lens of analogy-making-a topic of longtime interest in AI. We take our lead from a particular group of AI models that have emphasized the interplay between analogy-making and aesthetic sensibility. Central to the thinking behind these models is the idea of conceptual slippage, the process whereby one concept can “slip” to, or be replaced by, a related one, given sufficient contextual pressure. Extending this notion to perception and action, we argue that slippage and “seeing as” are central to both the creation and perception of artworks and other objects of design. We illustrate these points by drawing on a range of examples, both from computer models and from the real world. These observations suggest that a closer link should be established in AI between research on aesthetics, embodied cognition and perception, and analogy-making.
William W. York, Hamid R. Ekbia

Rationality {in|for|through} AI

Based on an assessment of the history and status quo of the concept of rationality within AI, I propose to establish research on (artificial) rationality as a research program in its own right, aiming at developing appropriate notions and theories of rationality suitable for the special needs and purposes of AI. I identify already existing initial attempts at and possible foundations of such an endeavor, give an account of motivations, expected consequences and rewards, and outline how such a program could be linked to efforts in other disciplines.
Tarek R. Besold

Usage of “Formal Rules” in Human Intelligence Investigations

In this paper, we challenge the anti-mechanism in the philosophy of mind and artificial intelligence by explaining that human intelligence, expertise, and skills acquaintance, relying mainly on the unconscious instinct could have been taken in formal rules. I do so, showing thereby the usage of the formal rules in unconscious processes for the study of human intelligence was raised in the second half of the 20 th century by psychoanalyst Matte Blanco. In this way, by exploring the bi-logic frameworkwith the general rules of unconscious processes, some positions that antimechanism has already taken should be re-examined in an interdisciplinary manner.
Tzu-Keng Fu

New Emergence as Supervenience Relieved of Problems

Supervenience and emergence are remarkable notions of cognitive science, which notably influenced especially philosophy of mind in the twentieth century. Issue of supervenient or emergent relationships is complicated, but it is possible to observe a tendency to prefer emergence at the expense of the notion of supervenience from nineties. This paper aims to answer question why it is so and is based on the fact that the answer might be interesting for artificial intelligence, which often uses the term of emergence. These two notions have always been very close to each other. This paper introduces development and common history of the concepts as well as changes of their relationship. The goal of the contribution is not only to consider the history of supervenience and emergence, but also to find appropriate distance to introduce thesis explaining the current relationship between them. The thesis could be simply formulated as follows: New use of the concept of emergence can be understood as a continuation of the idea of supervenience deprived of its fundamental problems.
Eliška Květová

Beyond Knowledge Systems

The central subject of the present paper is to analyze the step-by-step process leading from data-based to knowledge-based systems, and to show the relationship between knowledge systems and the structure and roles of human wisdom. In order to do this, we formulate our main question:What can be seen when focusing beyond the knowledge systems? We also present one of the acceptable answers.
Jozef Kelemen

Nature-Inspired Models


Selfish Genes and Evolutionary Computation

This paper deals with the relation between the so-called selfish genes and evolutionary computing without wishing to immerse into the biological evolution theories. The main goal is to show how a selfish gene could appear and how it is possible to demonstrate the presence of a selfish gene. We also want to answer the question if and how can the selfish gene be beneficial in the evolutionary computing.
Jan Zelinka

Nonlinear Trends in Modern Artificial Intelligence: A New Perspective

Artificial intelligent systems capable of learning, setting goals, solving problems, finding new solutions and unforeseen behavior scenarios without external assistance exist in a variety of odd models, each using different sets of assumptions and each having its own limitations. The author argues whether the modern artificial intelligent systems can be called truly intelligent so that to be able to realize the vast capabilities of the human brain. In order to generalize core issues of the main artificial intelligence domains such as fuzzy logics, probabilistic reasoning, bio-inspired techniques, neural networks apparatus, and neuroscience advances together with the new areas of chaos theory and nonlinear dynamics, a multidisciplinary analysis has been employed in the work. In the analysis, the evolution of a particular mathematical apparatus is being considered to justify the application of dynamic models with unstable dynamics used to solve intelligent problems of the next generation. The conclusion is soundly based on the idea that the future of artificial intelligence lies in the sphere of nonlinear dynamics and chaos that is absolutely critical to understanding and modeling cognition processes.
Elena N. Benderskaya

Membrane Computing in Robotics

This paper presents a new computational paradigm which can be successfully applied in robotics for the control of autonomous mobile robots. Membrane computing is a naturally parallel and distributed model of computation inspired by the structure and functioning of living cells. Numerical P systems, a type of membrane systems which operates with numerical values, and the extension, enzymatic numerical P systems, were used for modeling robot behaviors. Current results and developments of this innovative approach are also discussed and analyzed.
Ana Brânduşa Pavel, Cristian Ioan Vasile, Ioan Dumitrache

Implementing Enzymatic Numerical P Systems for AI Applications by Means of Graphic Processing Units

A P system represents a distributed and parallel computing model in which basic data structures are, for instance, multisets and strings. Enzymatic Numerical P Systems (ENPS) are a type of P systems whose basic data structures are sets of numerical variables. Separately, GPGPU (general-purpose computing on graphics processing units) is a novel technological paradigm which focuses on the development of tools for graphic cards to solve general purpose problems. This paper proposes an ENPS simulator based on GPUs and presents general concepts about its design and some future ideas and perspectives.
Manuel García-Quismondo, Luis F. Macías-Ramos, Mario J. Pérez-Jiménez

How to Design an Autonomous Creature Based on Original Artificial Life Approaches

We introduce new approaches for creating of autonomous agents. The life of such creatures is very similar to the animal’s life in the Nature, which learns autonomously from the simple tasks towards the more complex ones and is inspired by AI, Biology and Ethology. We present our established design of artificial creature, capable of learning from its experience in order to fulfill more complex tasks, which is based mainly on ethology. It integrates several types of action-selection mechanisms and learning into one system. The main advantages of the architecture is its autonomy, the ability to gain all information from the environment and decomposition of the decision space into the hierarchy of abstract actions, which dramatically reduces the total size of decision space. The agent learns how to exploit the environment continuously, where the learning of new abilities is driven by his physiology, autonomously created intentions, planner and neural network.
Pavel Nahodil, Jaroslav Vítků

When Artificial Becomes Natural


A View on Co-existence of Various Entities in Intelligent Environments

There is a number of traditional points of view from which intelligent environments are usually investigated. The most frequent among them are technological, social, economical, ethical, or political point of view. However, up to now there were just a few papers devoted to a research focused on co-existence of various entities that share such an intelligent environment and have to interact there. If we are viewing an intelligent environment as a collection (if not a community) of intelligent entities, capable of communication and performing activities based on a kind of mutual co-operation, then a kind of a co-existential point of view could be useful for further contemplations about various issues arising from this co-existence. The purpose of this paper is to present and discuss important aspects of co-existence of intelligent entities of various types (including humans) in intelligent environments and to formulate some interesting and important problems related to such a co-existence.
Peter Mikulecky, Petr Tucnik

Multi-agent Systems in Industry: Current Trends & Future Challenges

This paper introduces the multi-agent systems paradigm and presents some industrial applications of this AI approach, namely in manufacturing, handling and logistics domains. The road-blockers for the current weak adoption of this technology in industry are also discussed, and finally the current trends and several future challenges are pointed out to increase the wider dissemination and acceptance of the multi-agent technology in industry.
Paulo Leitao

Voice Conservation: Towards Creating a Speech-Aid System for Total Laryngectomees

This paper describes the initial experiments on voice conservation of patients with laryngeal cancer in an advanced stage. The final aim is to create a speechaid device which is able to “speak” with their former voices. Our initial work is focused on applicability of speech data from patients with an impaired vocal tract for the purposes of speech synthesis. Preliminary results indicate that appropriately selected synthesis method can successfully learn a new voice, even from speech data which is of a lower quality.
Zdeněk Hanzlíček, Jan Romportl, Jindřich Matoušek

Extended Mind: Is There Anything at All to Be Externalised?

The paper discusses Clark’s conception of extended mind and critically analyses his four criterions of externalised cognitive functions. Language as one of the most important means of externalisation is presented on the basis of Engelbart’s conception of human enhancement. Clark’s view of human-technology coupling is also strongly related to language which is regarded as a form of mind-transforming cognitive scaffolding. The material nature of language (as stressed by Clark in the domain of bodily gestures) is crucial for expression of mental concepts. This supports the belief that human cognitive enhancement is possible via technical means (e.g. AI-based speech prosthesis). However, from the philosophical point of view, even the categories of externality and internality of cognitive processes are very tricky.
Eva Zackova, Jan Romportl

Embodied Agent or Master of Puppets: Human in Relation with his Avatar

Neuronal and psychological processing of mental states of human operators of avatars is a relatively new but fast growing topic of interest. Most of the research focus on avatars meant as virtual agents controlled by humans, but the same explanations should also apply to other types, such as human-controlled robots. The most recent neuroscientific research proves that processing of avatar in-formation tends to increase activation in brain areas responsible for processing in-formation about body enhancements and embellishments. The author presents results of his behavioral studies and argues that the effect of treating an avatar as a body enhancement can bemediated by cognitive content such as attitudes, convictions and beliefs, causing different prospective behavior.
Mateusz Woźniak


Weitere Informationen

Premium Partner