L2szabályozás

L2szabályozás, also known as L2 regularization, is a technique used in machine learning to prevent overfitting by adding a penalty term to the loss function. This penalty is proportional to the sum of the squares of the model's weights. The goal of L2 regularization is to shrink the weights towards zero, but not necessarily to exactly zero. This encourages the model to use smaller weights, which generally leads to a simpler and more robust model that is less sensitive to individual data points.

The mathematical formulation of L2 regularization involves adding a term to the standard loss function. If the original loss function is denoted by $J(\theta)$, and the L2 regularization term is $\lambda \sum_{i=1}^{n} \theta_i^2$, where $\lambda$ is the regularization parameter and $\theta_i$ are the model's weights, then the regularized loss function becomes $J_{reg}(\theta) = J(\theta) + \lambda \sum_{i=1}^{n} \theta_i^2$. The parameter $\lambda$ controls the strength of the regularization. A higher $\lambda$ results in stronger regularization and smaller weights.

L2 regularization is widely used in various machine learning algorithms, including linear regression (often referred to as Ridge Regression), logistic regression, and neural networks. It helps to improve the generalization performance of models by reducing their variance. While it discourages large weights, it typically results in a model where all features have some non-zero weight, unlike L1 regularization which can drive some weights to exactly zero, effectively performing feature selection.