Systems and methods for mathematical regression with inexact feedback
The present invention provides methods for providing mathematical regression analysis. In particular, the method for conducting regression analysis comprises the steps of: selecting a regression model; selecting an initial set of regression parameters; applying the regression model to the initial set of regression parameters to create an initial set of regression values; selecting an improved set of regression values, wherein the improved set of regression values is selected from the set of initial regression values; generating a loss function based on the improved set; applying an iterative optimization method to the loss function and the improved set of regression values to generate a resultant set of regression values; and outputting the resultant set of regression values.
1. A method for conducting regression analysis comprising the steps of:
selecting a regression model;
selecting an initial set of regression parameters;
applying the regression model to the initial set of regression parameters to create an initial set of regression values;
receiving feedback;
selecting an improved set of regression values based on said feedback, wherein said improved set of regression values is changed from said initial set of regression values;
generating a loss function based on the improved set;
applying an iterative optimization method to the loss function and the improved set of regression values, wherein aid improved set of regression value is potentially changed from the previous iteration, to generate a resultant set of regression values; and
outputting the resultant set of regression values.
2. A computer program product comprising a computer readable memory storing computer executable instructions thereon that when executed by a computer perform the method of claim 1 .
3. A system for regression analysis, said system comprising a processor and a computer readable storage media comprising executable instructions thereon that when executed by said processor perform the method of claim 1 .
4. A method for conducting regression analysis comprising the steps of: a. selecting a regression model, regression parameters and a loss function; b. applying the regression model to the regression parameters to create output values; c. receiving feedback; d. updating output values and/or updating loss function based on said feedback, wherein said updating said output values comprises changing said output value, and wherein updating said loss function comprises replacing said loss function with a loss function having different attributes; e. determining if loss function indicates convergence, wherein if the loss function indicates convergence the regression analysis complete; and if the loss function indicates no convergence, said method further comprises: f. optimizing regression parameters and/or regression model based on loss function and output values to produce updated regression parameters and/or regression model, wherein steps b to e are reiterated until regression is complete;
wherein the optimizing is iterative.
5. The method of claim 4 , wherein the optimizing is by gradient descent.
6. The method of claim 4 , wherein optimizing utilizes the same method for each reiteration.
7. The method of claim 4 , wherein optimizing utilizes a different method for each reiteration.
8. A computer program product comprising a computer readable memory storing computer executable instructions thereon that when executed by a computer perform the method of claim 4 .
9. A system for regression analysis, said system comprising a processor and a computer readable storage media comprising executable instructions thereon that when executed by said processor perform the method of claim 4 .