Logistic log-loss function of one variable: Difference between revisions

Revision as of 17:53, 12 September 2014

Definition

The logistic log-loss function of one variable is obtained by composing the logarithmic cost function with the logistic function, and it is of importance in the analysis of logistic regression.

Explicitly, the function has the form:

$f (x) = - (p \ln (g (x)) + (1 - p) \ln (1 - g (x)))$

where $g$ is the logistic function and $\ln$ denotes the natural logarithm. Explicitly, $g (x) = \frac{1}{1 + e^{- x}}$ .

Note that $1 - g (x) = g (- x)$ , so the above can be written as:

$f (x) = - (p \ln (g (x)) + (1 - p) \ln (g (- x)))$

We restrict $p$ to the interval $[0, 1]$ . Conceptually, $p$ is the corresponding probability.

More explicitly, $f$ is the function:

$f (x) = p \ln (1 + e^{- x}) + (1 - p) \ln (1 + e^{x})$

@@ Line 2: / Line 2: @@
 The '''logistic log-loss function of one variable''' is obtained by composing the logarithmic cost function with the [[logistic function]], and it is of importance in the analysis of [[machinelearning:logistic regression|logistic regression]].
+Explicitly, the function has the form:
+<math>f(x) = -(p \ln (g(x)) + (1 - p) \ln(1 - g(x)))</math>
+where <math>g</math> is the [[logistic function]] and <math>\ln</math> denotes the natural logarithm. Explicitly, <math>g(x) = \frac{1}{1 + e^{-x}}</math>.
+Note that <math>1 - g(x) = g(-x)</math>, so the above can be written as:
+<math>f(x) = -(p \ln (g(x)) + (1 - p) \ln (g(-x)))</math>
+We restrict <math>p</math> to the interval <math>[0,1]</math>. Conceptually, <math>p</math> is the corresponding probability.
+More explicitly, <math>f</math> is the function:
+<math>f(x) = p \ln(1 + e^{-x}) + (1 - p) \ln (1 + e^x)</math>