Antenna array technology can be applied to exploit the spatial diversity of digital wireless communications and to increase the channel capacity. Signal (waveform) estimation in most array processing algorithms is bas...
详细信息
A transaction management model for distributed informationsystems that utilizes actor-like objects (TrActor) which deviate from conventional actors in their objects types, messages types, and mail queue is described....
详细信息
This paper presents a system architecture for autonomous navigation for a mobile platform designed on the view of activities. A control unit that acts upon information posted on a blackboard supervises these activitie...
详细信息
One important feature of neural networks is that their basic processing elements compute nonlinear functions such as sigmoidal or linear threshold functions. While most existing mathematical tools are effective in ana...
详细信息
One important feature of neural networks is that their basic processing elements compute nonlinear functions such as sigmoidal or linear threshold functions. While most existing mathematical tools are effective in ana...
详细信息
ISBN:
(纸本)0780305574
One important feature of neural networks is that their basic processing elements compute nonlinear functions such as sigmoidal or linear threshold functions. While most existing mathematical tools are effective in analyzing problems with linear structures, they are inadequate in dealing with nonlinear problems. We develop novel techniques based on classical tools such as rational approximation and harmonic analysis to study the computational properties of neural networks. Using such techniques, we can characterize the class of functions whose complexity is almost the same among various models of neural networks with feedforward structures. As a consequence of this characterization, for example, we prove that any dept.-(d + 1) network of sigmoidal units computing the parity function of n inputs must have Ω(dn1/d-∈) units, for any fixed ∈ > 0. This lower bound is almost tight since we can compute the parity function with 0(dn1/d) sigmoidal units in a dept.-(d +1) network. Our techniques also generalize to networks whose elements can be approximated by piecewise low degree rational functions. These almost tight bounds are the first known complexity results on the size of neural networks computing Boolean functions with continuous-output elements and with dept. more than two.
The importance of production systems in artificial intelligence (AI) has been repeatedly demonstrated by a large number of expert systems. As the number and size of expert systems grow, there has however been an emerg...
详细信息
The importance of production systems in artificial intelligence (AI) has been repeatedly demonstrated by a large number of expert systems. As the number and size of expert systems grow, there has however been an emerging obstacle in such AI applications: the large processing time. The need for faster execution of production systems has spurred research in both the software and hardware domains, including connectionist architectures. This paper surveys various aspects of parallel distributed processing of production systems. Approaches taken to date to solve the problems associated with production systems are classified here into three levels: the algorithmic level, the parallel implementation level, and the connectionist level. Several pattern matchers and multiple rule firing principles are presented to demonstrate the algorithm level improvement. Several parallel implementation efforts are surveyed along with experimental results on real machines or with simulators. The presentation of three different types of connectionist production systems (local, distributed, and hierarchical representation) completes this survey. Finally, we explore some potential avenues towards the implementation of a true asynchronous parallel production system.
This paper focuses on the task of design verification using both knowledge of the structure of a device and its intended functions. In particular, it addresses the question of when one can say a behavior predicted by ...
We extend the results of our previous paper [8] to the nonlinear case: The Lloyd polynomial of the covering has at least R distinct roots among 1,…, n, where R is the covering radius. We investigate PWC with diameter...
详细信息
暂无评论