Computability and Learnability of Weightless Neural Networks
نویسنده
چکیده
The aim of this paper is to compare what can be computed with what can be learnable by Artificial Neural Networks (ANN). We will approach the learnability problem in ANN by fixing a particular class of ANN: the weightless neural networks (WNN) ([4)) and restricting ourselves to a particular learning paradigm: learning to recognise sentences of a formallanguage. We will approach the computability problem by comparing the class of languages that can be recognised by WNN with the classes of languages in the Chomsky hierarchy.
منابع مشابه
Probabilistic automata simulation with single layer weightless neural networks
Computability of weightless neural networks is the major topic of this paper. In previous works it has been shown that, one can simulate a Turing machine with a weightless neural network (WNN) with an infinite tape. And it has also been shown that one can simulate probabilistic automata with a WNN with two queues. In this paper, we will show that is possible to simulate a probabilistic automata...
متن کاملLearning with recurrent neural networks
This thesis examines so-called folding neural networks as a mechanism for machine learning. Folding networks form a generalization of partial recurrent neural networks such that they are able to deal with tree structured inputs instead of simple linear lists. In particular, they can handle classical formulas { they were proposed originally for this purpose. After a short explanation of the neur...
متن کاملTridimensional Pattern Reconstruction by Using Weightless Artificial Neural Networks
1. ABSTRACT This paper describes the structure and behavior of a system, composed by a set of weightless artificial neural networks, which is capable of learning different images and then reconstructing an image according to the closest learned pattern. This paper presents a technique which considers Hamming distance for pattern learning and reconstruction, therefore it is posible to study the ...
متن کاملWANN-Tagger - A Weightless Artificial Neural Network Tagger for the Portuguese Language
Weightless Artificial Neural Networks have proved to be a promising paradigm for classification tasks. This work introduces the WANN-Tagger, which makes use of weightless artificial neural networks for labelling Portuguese sentences, tagging each of its terms with its respective part-of-speech. A first experimental evaluation using the CETENFolha corpus indicates the usefulness of this paradigm...
متن کاملImplementation of Probabilistic Automata in Weightless Neural Networks
The objective of this paper is to analyze the practical viability of the results theoretical concerning the relationship of a class of weightless neural networks, known as General Single-layer Sequential Weightless Neural Networks (GSSWNNs), and Probabilistic Automata (PA). This study was based on the theoretical model development by de Souto [1]. This model shows the computational equivalence ...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
عنوان ژورنال:
دوره شماره
صفحات -
تاریخ انتشار 2013