Title :
An optimal dimension expansion procedure for obtaining linearly separable subsets
Author :
Tseng, Yuen-Hsien ; Wu, Ja-Ling
Author_Institution :
Dept. of Comput. Sci. & Inf. Eng., Nat. Taiwan Univ., Taipei, Taiwan
Abstract :
The authors study the necessary and sufficient condition for linearly separable subsets and then propose an optimal dimension expansion procedure that makes any mapping to be performed by perceptrons learnable by an error-correction procedure. For n-bit parity check problems, it is shown that only one additional dimension is augmented to make them solvable by single-layer perceptrons. Other applications such as for decoding error-correcting codes are also considered
Keywords :
decoding; error correction; learning systems; neural nets; set theory; decoding; error-correcting codes; error-correction; learning systems; linearly separable subsets; necessary and sufficient condition; neural nets; optimal dimension expansion procedure; perceptrons; set theory; Computer errors; Computer science; Decoding; Digital circuits; Equations; Error correction codes; Modems; Parity check codes; Sufficient conditions;
Conference_Titel :
Neural Networks, 1991. 1991 IEEE International Joint Conference on
Print_ISBN :
0-7803-0227-3
DOI :
10.1109/IJCNN.1991.170758
