Search

In this paper we show how symbolic Artificial Intelligence (AI) techniques can be used to develop intelligent Virtual Reality (VR) environments. We have developed an expert system architecture for building intelligent agents that respond to voice and gesture commands in virtual environments. To demonstrate the utility of this we present a simple application that allows a user to drive a virtual robot (the VirBot) around a virtual environment with multimodal commands. This software agent is able to accept spoken commands from the user, interpret them using natural language understanding and interact with and modify the virtual environment. Command recognition is further improved by the use of contextual information which reduces recognition error rates by 31%. Context is encoded using scripts, making it easy to compose inferences about events for which there is incomplete or fuzzy information.

A multi-agent system architecture that facilitates the design, implementation and communication of autonomous dedicated agents is discussed. The architecture allows the learning abilities of agents using both connectionist and symbolic artificial intelligence (AI) techniques to be investigated. Identifying a particular supervisor for a student is a very time-consuming task. Processing all the information related to students and staff members, distributing this information, obtaining feedback from the participants, making sure that all students have supervisors and that all members of staff have similar numbers of students to supervise can be a very tedious task. The Project Monitoring Intelligent Agent (PMIA) system has been developed with the aim of automating as much of this process as possible. PMIA is a multi-agent based tool designed using the philosophy of Internet software agents and uses a hybrid architecture that combines the advantages of both symbolic and connectionist approaches to artificial intelligence problem solving. The system design is influenced by the agent architecture proposed by the Knowledge Sharing Effort using the Knowledge, Query and Manipulation Language for agent communication; this is an Internet agent based system, designed using HTML, Perl and Visual C++, that requires a WWW server and the help of Web browsers.

The brain uses recurrent spiking neural networks for higher cognitive functions such as symbolic computations, in particular, mathematical computations. We review the current state of research on spike-based symbolic computations of this type. In addition, we present new results which show that surprisingly small spiking neural networks can perform symbolic computations on bit sequences and numbers and even learn such computations using a biologically plausible learning rule. The resulting networks operate in a rather low firing rate regime, where they could not simply emulate artificial neural networks by encoding continuous values through firing rates. Thus, we propose here a new paradigm for symbolic computation in neural networks that provides concrete hypotheses about the organization of symbolic computations in the brain. The employed spike-based network models are the basis for drastically more energy-efficient computer hardware – neuromorphic hardware. Hence, our results can be seen as creating a bridge from symbolic artificial intelligence to energy-efficient implementation in spike-based neuromorphic hardware.

There is a natural question the reader of Cherniavsky's publications of the last years may ask: Are there limitations of V.S. Cherniavsky? Cherniavsky's question is still, "whether artificial intelligence is limited in the sense of being separated from intelligence of the human by some natural boundary." He answers it roughly this way: if we assume it is not, we get into trouble. The `Main Assumption' of artificial intelligence is that human thinking is as deterministic as machines are:Given some text t and some evidence i (of t), a deterministic device can (after some time) judge it "true" eventually.

Winston and Horn's Lisp is a textbook on symbolic computation and the Lisp language. It was originally published in 1981, based on a set of course notes developed for the introductory course in artificial intelligence at MIT.

This article describes the features of artificial intelligence problems that have affected the design of artificial intelligence languages. It describes ways in which artificial intelligence languages have been developed to cope with these features. It explains the role of the programming environment in which the artificial intelligence researcher works and how it is closely coupled with the languages he is using. Finally, some lessons that software engineering can draw from the artificial intelligence experience in using such environments are described.

Intelligence evolves out of matter, so said the Sankhya philosophers of ancient India. The discipline of artificial intelligence (Al), which was established some 30 years ago, has confirmed the validity of the above assertion. Recently, a number of AI applications have been successfully demonstrated, generating a great deal of excitement and interest in scientific and technical circles. In this special issue of Optical Engineering a representative set of applications that incorporate Al principles is presented.

Here is a volume that offers concrete examples and exercise to illustrate a wide variety of recently developed techniques used in implementing modern Artificial Intelligence programs, and describes sophisticated LISP concepts and applied them to real AI problems.

After Fortran, Lisp is the second-oldest high-level programming language in widespread use. Strongly associated with artificial intelligence, its powerful facilities for symbolic processing and incremental development can be of benefit across a wide range of computer applications

IT is one of the objections to the Spearman two-factor theory1 that the measure of general intelligence, g is not necessarily invariant for linear transformations of the test scores. I have not followed the discussion on this point at all closely, but a remark interpolated during the joint discussion at Leicester of Sections A and J of the British Association is perhaps worth repeating here, if only to be answered.

This book gives an excellent overview of heuristic programs in the field of artificial intelligence. The author describes oever 30 programs which "are doing things often called intelligent when done by humans." The book is divided into 13 chapters and each contains exercises and a bibliography.

Probably the ultimate achievement in simulating human operation is the creation of an artificial intelligence. At present, artificial-intelligence research has revealed aspects of human ignorance, particularly in the fields of linguistics and psychology. In the future, it is desired to produce machines which can translate languages, recognise printed letters and, most important of all, possess a capacity for self improvement.