Chartered Financial Analyst (CFA) Practice Exam Level 2

What aspect of machine learning does "overfitting" relate to?

1. Training cycles always improve predictions

2. Training cycles might not improve predictions

3. Training cycles are irrelevant to model performance

4. Training cycles can lead to underfitting

The correct answer is: Training cycles might not improve predictions

Overfitting is a concept in machine learning that occurs when a model learns the training data too well, capturing noise and outliers instead of the general patterns of data. This results in a model that performs excellently on the training dataset but fails to generalize to new, unseen data, illustrating that prediction capability might not consistently improve with additional training cycles. Consequently, while initial training cycles may enhance model performance, excessive training can degrade the model's ability to make accurate predictions on new data, highlighting the relationship between model training cycles and potential prediction issues. In this context, the correct choice recognizes that training cycles may not always yield improvements in predictions, especially when overfitting is a factor influencing model performance.