Natural Limits of Artiicial Realities
نویسندگان
چکیده
Without hands and feet He goes fast and grasps; without eyes He sees; without ears He ears. He knows whatever is to be known, yet there is none who knows Him. They say He is the foremost, the great Innnite Being. \S'vetas'vatara Upanishad" The universe is information and we are stationary in it, not three-dimensional and not in space or time. The information fed to us we hypostatize into the phenomenal world. Abstract We propose a preliminary exploration of the outskirts of the research area concerned with the design of artiicial reality systems. Our goal is to clarify the concept of artiicial reality and its connections both with other kinds of computer systems and with the world we live in (natural reality). In order to provide a framework for such a discussion, we sketch a list of dimensions along which to classify artiicial realities. We would like to thank Enrico Franconi (for telling us of the Conference), Enzo Minervini (for pointing out the relevance of Bishop Berkeley), Giorgio Satta (for helping with L a T E X), Sandy Stone (the kindest presence in the matrix) and Achille C. Varzi (for his valuable comments on a preliminary draft).
منابع مشابه
Can Zipf Analyses and Entropy Distinguish between Artiicial and Natural Language Texts?
We study statistical properties of natural texts written in English and of two types of artiicial texts. As statistical tools we use the conventional and the inverse Zipf analyses, the Shannon entropy and a quantity which is a nonlinear function of the word frequencies , the frequency relative \entropy". Our results obtained by investigating eight complete books and sixteen related artiicial te...
متن کاملNatural Limits of Arti cial Realities
We propose a preliminary exploration of the outskirts of the research area concerned with the design of arti cial reality systems. Our goal is to clarify the concept of arti cial reality and its connections both with other kinds of computer systems and with the world we live in (natural reality). In order to provide a framework for such a discussion, we sketch a list of dimensions along which t...
متن کاملReconngurable Acousto-optic Processor for Artiicial Neural Network Training
We introduce the initial experiments of a project to train large-scale artiicial neural networks on a reconngurable optical computing platform, based on a calomel (mercurous chloride) acousto-optic matrix-vector multiplier. In this paper we present the results of implementing learning on a perceptron with bipolar outputs. These experiments are intended to address the major diiculties anticipate...
متن کاملArtiicial Neural Networks and Artiicial Immune Systems: Similarities and Diierences
Both the nervous system and the immune system are complex biological systems. Recognition and cat-egorization are the major functions of both systems. The information processing principles of these natural systems inspired in developing intelligent problem-solving techniques, namely, the Artiicial Neural Network (ANN) and the Artiicial Immune System (AIS). Though ANNs are well established techn...
متن کاملOo-line Learning of a Domain-speciic Transmission Function for Artiicial Evolution 1 Motivation 1.1 Artiicial vs. Natural Evolution
Sustained progress in artiicial evolution is dependent on the ability of the variation process{via the transmission function 1], to continue to generate individuals of higher tness. However, the design of a transmission function to yield good performance is a diicult task, and becomes increasingly problematic in complex, open-ended domains such as those explored in the eld of Ar-tiicial Life. I...
متن کامل