lecture 2 Flashcards
What is the basic recipe for machine learning?
Abstract the problem, choose instances and features, choose a model class, search for a good model.
What are the common types of machine learning tasks?
Classification, Regression, Clustering, Density estimation, Generative Modeling, Online learning, Reinforcement Learning, Structured Output Learning.
What is regression in machine learning?
A method to predict a continuous target variable based on one or more features.
What is the difference between classification and regression?
Classification predicts discrete labels, while regression predicts continuous values.
What is the purpose of a loss function in machine learning?
To quantify the difference between predicted values and actual values, guiding model optimization.
What is the primary search method used in machine learning optimization?
Gradient descent.
What is the equation of a simple linear regression model?
f(x) = wx + b
What do w and b represent in linear regression?
w is the weight (slope), and b is the bias (intercept).
What happens if w is positive in a linear regression model?
The function increases as x increases.
What happens if w is negative in a linear regression model?
The function decreases as x increases.
What is the goal of linear regression?
To find the best-fitting line that minimizes the loss function.
What is a feature in machine learning?
An individual measurable property or characteristic of the data.
What is a target variable?
The variable that the model is trying to predict.
What is a dataset instance?
A single data point consisting of features and a target value.
What is the dot product in linear regression?
The sum of the products of corresponding elements in two vectors.
How does increasing the number of features affect a linear model?
It increases model complexity and may require additional optimization techniques.
What is the name of the method used to find the best values for w and b?
Gradient Descent.
What is overfitting in machine learning?
When a model learns noise in the training data instead of generalizing to new data.
What is underfitting in machine learning?
When a model is too simple to capture patterns in the data.
What is the difference between supervised and unsupervised learning?
Supervised learning has labeled data, while unsupervised learning does not.
What is a weight in a machine learning model?
A parameter that determines the importance of a feature in predicting the target.
What is the effect of a high learning rate in gradient descent?
Faster convergence but potential overshooting of the optimal solution.
What is the effect of a low learning rate in gradient descent?
Slow convergence but more precise optimization.
What does a high bias indicate in a machine learning model?
The model makes consistent errors and is too simple (underfitting).
What does a high variance indicate in a machine learning model?
The model is too sensitive to training data and does not generalize well (overfitting).
What is the difference between batch and stochastic gradient descent?
Batch gradient descent uses all data at once, while stochastic updates the model per data point.
What is the role of the intercept (b) in a linear regression model?
It determines where the regression line crosses the y-axis.
What is meant by normalizing features in machine learning?
Rescaling features to have a similar range to improve model performance.
What is a loss function used for in machine learning?
To measure how well a model’s predictions match the actual values.
What is a hyperparameter in machine learning?
A configuration value that is set before training (e.g., learning rate, number of layers).
