Machine Learning Fundamentals: Bias and Variance

These flashcards cover key concepts related to bias and variance in machine learning, including model fit, overfitting, and methods to improve model performance.

What is the key difference between bias and variance in machine learning models?

Bias refers to the error due to overly simplistic assumptions in the learning algorithm, while variance refers to the error due to excessive complexity in the model.

What happens to a model with low bias and high variance?

It may fit the training data well but perform poorly on unseen data due to overfitting.

What is the goal of finding the 'sweet spot' in machine learning models?

To balance model simplicity and complexity, achieving low bias and low variance.

What is overfitting in machine learning?

When a model fits the training data too closely and performs poorly on testing data.

What are the three commonly used methods for finding the sweet spot between simple and complicated models?

Regularization, boosting, and bagging.

What is linear regression, according to the lecture?

A machine learning method that fits a straight line to the training data.

Why might a squiggly line model have low bias but high variance?

Because it fits the training data very closely but struggles with different datasets.

What is the significance of the sums of squares in evaluating model fit?

It measures the distances from the fitted lines to the data points to determine how well the model predicts outcomes.

In the context of the mice weight and height example, what does a 'training set' refer to?

The subset of data used to fit the machine learning model.

What does 'testing set' mean in machine learning?

The subset of data used to assess the performance of the machine learning model.