Logistic regression models are trained using the same process as linear regression models, with two key distinctions:
- Logistic regression models use Log Loss as the loss function instead of squared loss.
- Applying regularization is critical to prevent overfitting.
The following sections discuss these two considerations in more depth.
Log Loss
In the Linear regression module, you used squared loss (also called L2 loss) as the loss function. Squared loss works well for a linear model where the rate of change of the output values is constant. For example, given the linear model $y' = b + 3x_1$, each time you increment the input value $x_1$ by 1, the output value $y'$ increases by 3.
However, the rate of change of a logistic regression model is not constant. As you saw in Calculating a probability, the sigmoid curve is s-shaped rather than linear. When the log-odds ($z$) value is closer to 0, small increases in $z$ result in much larger changes to $y$ than when $z$ is a large positive or negative number. The following table shows the sigmoid function's output for input values from 5 to 10, as well as the corresponding precision required to capture the differences in the results.
| input | logistic output | required digits of precision |
|---|---|---|
| 5 | 0.993 | 3 |
| 6 | 0.997 | 3 |
| 7 | 0.999 | 3 |
| 8 | 0.9997 | 4 |
| 9 | 0.9999 | 4 |
| 10 | 0.99998 | 5 |
If you used squared loss to calculate errors for the sigmoid function, as the
output got closer and closer to 0 and 1, you would need more memory to
preserve the precision needed to track these values.
Instead, the loss function for logistic regression is Log Loss. The Log Loss equation returns the logarithm of the magnitude of the change, rather than just the distance from data to prediction. Log Loss is calculated as follows:
\(\text{Log Loss} = \sum_{(x,y)\in D} -y\log(y') - (1 - y)\log(1 - y')\)