Digital Commons Network^™

Using Discovery-Based Learning To Prove The Behavior Of An Autonomous Agent, David N. Mezera

Theses and Dissertations

Computer-generated autonomous agents in simulation often behave predictably and unrealistically. These characteristics make them easy to spot and exploit by human participants in the simulation, when we would prefer the behavior of the agent to be indistinguishable from human behavior. An improvement in behavior might be possible by enlarging the library of responses, giving the agent a richer assortment of tactics to employ during a combat scenario. Machine learning offers an exciting alternative to constructing additional responses by hand by instead allowing the system to improve its own performance with experience. This thesis presents NOSTRUM, a discovery-based learning DBL system …

Procase: A Prototype Of Intelligent Case-Based Process Planning System With Simulation Environment, Hao Yang, Wen F. Lu

Mechanical and Aerospace Engineering Faculty Research & Creative Works

An intelligent case-based process planning system with interactive graphic simulation environment, PROCASE, is developed lo demonstrate an integrated methodology of case-based process planning system. In PROCASE, both the mechanical part features and the machining operations are represented with a frame-based scheme. PROCASE contains a retriever, a modifier, a simulator and a repairer. It distinguishes itself from traditional rule-based process planning systems by representing the process planning knowledge through previous process planning cases instead of production rules. It therefore can overcome some problems in the traditional rule-based expert systems. PROCASE currently resides in IRIS Indigo workstation. With a user-friendly graphic environment, …

Learning In Connectionist Networks Using The Alopex Algorithm, Kootala Pattath Venugopal

Electronic Theses and Dissertations 1990 - 1999

The Alopex algorithm is presented as a universal learning algorithm for connectionist models. It is shown that the Alopex procedure could be used efficiently as a supervised learning algorithm for such models. The algorithm is demonstrated successfully on a variety of network architectures. Such architectures include multilayer perceptrons, time-delay models, asymmetric, fully recurrent networks and memory neuron networks. The learning performance as well as the generation capability of the Alopex algorithm are compared with those of the backpropagation procedure, concerning a number of benchmark problems, and it is shown that the Alopex has specific advantages over the backpropagation. Two new …

Learning Algorithms For Artificial Neural Networks: A Closer Look At Back Propagation, Simplex And Monte Carlo, Sergio Alberto Cedeño

Graduate Theses and Dissertations

No abstract provided.

Real Time Maze Traversal, Robert Spina

Theses

This thesis investigates the application of artificial intelligence to real time control in combination with image analysis techniques. The system consists of a "mouse" and a "cheese" in a two dimensional maze which are viewed by a video camera connected to a frame grabber board in a personal computer (PC). The maze is designed to be alterable during the mouse's navigation towards the "cheese" or target. The pc runs the frame grabber software which analyzes the maze to find the cheese, mouse and movement options during traversal. Parameters about the mouse's position, the maze constraints in the vicinity of the …

Intelligent Knowledge Acquisition System, Bong-Soo Youn

Theses

IKAS (Intelligent Knowledge Acquisition System) is an expert system building shell with a machine-assisted concept checker for easy integrated rule-entry. Unlike most expert system shells currently available, IKAS not only performs rule syntax checking, but also maintains semantic integration of the knowledge base by using a Concept Base. The Concept Base is incrementally grown during the cycle of expert system application development and safeguards the whole domain from conceptually conflicting rules by finding possible inconsistencies or duplications of concepts and giving recommendations for possible solutions. Also automatic generalized concept forming using the Inductive Extension Generalization method gives IKAS a capability …

Learning System, Bhavesh T. Adani

Theses

Many rule-based learning systems have been implemented which generate and update rules/facts from illustrative examples provided by the user. But very few of them have succeeded in achieving the efficiency and accuracy of an ideal system.

The approach made here is to continuously update the hierarchical structure of rules/facts of a knowledge-base so that efficient and accurate solutions to any query can be achieved easily and immediately from the many possible solutions to the query. As the rule-based systems deal with a very large knowledge-base, it is sometimes necessary to apply a learning technique that finds optimum solutions efficiently.

The …

Artificial Intelligence: Myths And Realities, Hugo D'Alarcao

Bridgewater Review

Artificial intelligence the name conjures images of mechanical monsters, the Golem, Dr. Frankenstein’s creation and the rebellious computer Hal. We have always been fascinated by the possibility of creating a machine in our image, but this fascination is often accompanied by apprehension. We fear losing control of our creation and suspect that it might turn against us. It is this duality, this conflict between the desire to create and the fear of the consequences of the creation that has been so successfully exploited by writers. It is also, in part, this fascination that has recently brought the field of Artificial …