CS325 Topic 7, Lectures 1,2,3,4

The study of computer algorithms that improve automatically through experience and by the use of data.

A specification of a mathematical or probabilistic relationship between different variables in a dataset.

A modeling error where the model performs well on the training data but poorly on new unseen data.

A modeling error where the model does not perform well on the training data, leading to poor generalization.

The difference between the average prediction of the model and the true value, often leading to oversimplification and high prediction error.

The variability of model predictions for a given dataset, which leads to poor performance on new data.

A method used to evaluate the performance of a classification model by measuring the correctness of its predictions.

It refers to correctly predicting a positive class, such as identifying a sick person who is actually sick.

It refers to incorrectly predicting a positive class, like diagnosing a healthy person as sick.

It is a classifier that predicts new data points based on the closest neighbors in an existing dataset.

As the number of features increases, modeling error increases exponentially, making the algorithm less effective.

A machine learning algorithm that classifies categorical outcomes based on conditional probability, assuming all features are independent.

An algorithm that predicts a continuous numerical value based on the relationship between one input and one output variable.

The output variable whose value is predicted based on the independent variable.

The input variable that explains or predicts changes in the dependent variable.

Each card contains a concept and its explanation to help with studying these machine learning topics!