Ch 2 Flashcards
What is the loss function used in 1D linear regression?
Least squares loss function
The least squares loss function measures the average squared difference between predicted and actual values.
What is the goal of supervised learning?
Mapping from one or more inputs to one or more outputs
Supervised learning involves training a model using labeled data.
What is the purpose of the ‘predict’ function in the Python implementation?
To compute predictions: y = m * x + b
This function predicts output values based on input features and model parameters.
What does the Mean Squared Error (MSE) loss function measure?
Quantifies the error between predicted outputs and true outputs
MSE is calculated as the average of the squares of the errors.
What is the primary objective of the training process in supervised learning?
To develop a model that accurately represents the underlying patterns of the training data.
Fill in the blank: The technique used for updating model parameters in supervised learning is called _______.
gradient descent.
What is overfitting in the context of supervised learning?
When a model is too complex and fits to statistical peculiarities of data.
What distinguishes training from inference in a supervised learning model?
Training develops the model; inference uses the model to make predictions.
What is the definition of a derivative in the context of a function?
The rate of change of the output of the function with respect to its input at that point.
What are the components of a 1D linear regression model?
Parameters: y-offset, slope
The y-offset is the intercept, and the slope determines the steepness of the line.
True or False: The training process can be executed rapidly compared to inference.
False.
What type of data is used for testing a supervised learning model?
Separate test dataset of input/output pairs.
What does a loss function measure in supervised learning?
How bad the model is at mapping inputs to outputs.
What is the relationship assumption in 1D linear regression?
The relationship between input and output is a straight line.
What is the primary role of parameters in a supervised learning model?
Parameters affect the outcome of the equation.
What is the notation for inputs and outputs in a supervised learning model?
Inputs: Roman letters; Outputs: Normal letters.
Fill in the blank: In regression tasks, we predict _______ values.
continuous.
