You are training a spam classifier. You notice that you are overfitting the training data. Which three actions can you take to resolve this problem? (Choose three.)

More training examples expose the model to a wider variety of patterns, reducing its ability to memorize specific training instances and improving generalization to unseen data.

Reducing training examples worsens overfitting by giving the model fewer diverse examples, making it easier to memorize the smaller dataset rather than learn generalizable patterns.

Reducing the feature set lowers model complexity and eliminates noisy or irrelevant signals that the model might overfit to, forcing it to rely on the most predictive features.

Adding more features increases model dimensionality and complexity, providing additional dimensions along which the model can overfit the training data.

Increasing regularization parameters (such as L1 or L2 lambda) penalizes large coefficient values, constraining model complexity and preventing the model from fitting noise in the training data.

Decreasing regularization parameters weakens the complexity penalty, allowing the model to assign larger weights to individual features and fit the training data more aggressively.