1 February 1990 A Neural Network For L1 Norm Linear Regression
Author Affiliations +
Proceedings Volume 1196, Intelligent Control and Adaptive Systems; (1990) https://doi.org/10.1117/12.969918
Event: 1989 Symposium on Visual Communications, Image Processing, and Intelligent Robotics Systems, 1989, Philadelphia, PA, United States
Abstract
We are presenting a neural network optimization circuit for robust regression. The minimization of Least Square Error (LSE), L2 norm, cost functions is predominantly used for fitting functions to data points, but LSE approaches are highly sensitive to outlier points, or points that don't follow the trend. To alleviate this problem, we replace the L2 norm error function by an L1 norm error function which sums the absolute deviation of the errors and puts less weight on outlier points. An analog neural network optimization circuit is then developed to minimize the sum of the L1 norm error function. Simulation examples are presented on example data sets that compare the neural network solution with LSE solution .
© (1990) COPYRIGHT Society of Photo-Optical Instrumentation Engineers (SPIE). Downloading of the abstract is permitted for personal use only.
James P. Helferty, James P. Helferty, James A. Stover, James A. Stover, John J. Helferty, John J. Helferty, } "A Neural Network For L1 Norm Linear Regression", Proc. SPIE 1196, Intelligent Control and Adaptive Systems, (1 February 1990); doi: 10.1117/12.969918; https://doi.org/10.1117/12.969918
PROCEEDINGS
12 PAGES


SHARE
Back to Top