Numeric Law Discovery using Neural Networks
نویسندگان
چکیده
This paper proposes a new connectionist approach to numeric law discovery; i.e., neural networks (law-candidates) are trained by using a newly invented second-order learning algorithm based on a quasi-Newton method, called BPQ, and the MDL criterion selects the most suitable from law-candidates. The main advantage of our method over previous work of symbolic or connectionist approach is that it can efficiently discover numeric laws whose power values are not restricted to integers. Experiments showed that the proposed method works well in discovering such laws even from data containing irrelevant variables or a small amount of noise.
منابع مشابه
Law discovery from financial data using neural networks
This paper describes an experimental study for discovering underlying laws of market capitalization using BS (Balance Sheet) items. For this purpose, we apply law discovery methods based on neural networks: RF5 (Rule Finder) discovers a single numeric law from data containing only numeric values, RF6 discovers a set of nominally conditioned polynomials from data containing both nominal and nume...
متن کاملLaw Discovery using Neural Networks
This paper proposes a new connectionist approach to numeric law discovery; i.e., neural networks (law-candidates) are trained by using a newly invented second-order learning algorithm based on a quasi-Newton method, called BPQ, and the Minimum Description Length criterion selects the most suitable from lawcandidates. The main advantage of our method over previous work of symbolic or connectioni...
متن کاملDynamic Sliding Mode Control of Nonlinear Systems Using Neural Networks
Dynamic sliding mode control (DSMC) of nonlinear systems using neural networks is proposed. In DSMC the chattering is removed due to the integrator which is placed before the input control signal of the plant. However, in DSMC the augmented system is one dimension bigger than the actual system i.e. the states number of augmented system is more than the actual system and then to control of such ...
متن کاملTwo ways of using artificial neural networks in knowledge discovery from chemical materials data
In the application area of chemical materials, data mining methods have been used for more than a decade. By far most popular have from the very beginning been methods based on artificial neural networks. However, they are frequently used without awareness of the difference between the numeric nature of knowledge obtained from data by neural network regression, and the symbolic nature of knowle...
متن کاملStabilization of Nonlinear Control Systems through Using Zobov’s Theorem and Neural Networks
Zobov’s Theorem is one of the theorems which indicate the conditions for the stability of a nonlinear system with specific attraction region. We have applied neural networks to approximate some functions mentioned in Zobov’s theorem in order to find the controller of a nonlinear controlled system whose law in a mathematical manner is difficult to make. Finally, the effectiveness and the applica...
متن کامل