Discovery of a Set of Nominally Conditioned Polynomials

: This paper shows that a connectionist law discovery method called RF5X can discover a law in the form of a set of nominally conditioned polynomials, from data containing both nominal and numeric values. RF5X learns a compound of nominally conditioned polynomials by using single neural networks, and selects the best one among candidate networks, and decomposes the selected network into a set of rules. Here a rule means a nominally conditioned polynomial. Experiments showed that RF5X works well in discovering such a law even from data containing irrelevant variables and a small amount of noise. 1 T = kr3/2 BACON [3] [5, 9] BPQ[8] RF5 [7] RF5 [4] RF5 F = 4π q1q2/r2 r q1 q2 F F = 8897.352 q1q2/r2

• Type: Conference
• Year: 1999
• Where: DIS
• Authors: Ryohei Nakano, Kazumi Saito
• Keywords: Connectionist Law Discovery, DIS 1999, Method Called Rf5x, Single Neural Networks, Theoretical Computer Science