Flashcards · Machine Learning · Free
Machine Learning flashcards, generated for you.
Example Machine Learning study cards to learn from right now — then generate a full set from your own notes (plus a practice quiz) and export to Quizlet or Anki. Free, no account needed.
Example Machine Learning flashcards
What is machine learning in simple terms?
Teaching computers to learn patterns from data and make predictions without being explicitly programmed for every scenario.
What's the difference between supervised and unsupervised learning?
Supervised: you provide labeled examples (input + correct answer) so the model learns to predict. Unsupervised: you provide unlabeled data and the model finds hidden patterns on its own.
What does a 'feature' mean in machine learning?
A feature is a single measurable input variable (e.g., house size, age, color) that the model uses to make predictions.
Why do we split data into training and testing sets?
Training set teaches the model patterns; testing set checks if it learned *generalizable* patterns or just memorized (overfitting). This ensures real-world performance.
What is overfitting and why is it bad?
Overfitting occurs when a model memorizes training data noise instead of learning true patterns, causing poor performance on new data. High train accuracy but low test accuracy is a sign.
Define the loss function and explain its role.
A loss function measures how wrong predictions are (e.g., squared error). The model minimizes this during training to find parameters that produce better predictions.
What does 'gradient descent' intuitively do?
Iteratively adjusts model parameters by taking small steps in the direction that reduces loss, like walking downhill to find the valley (minimum loss).
What is regularization and why use it?
Regularization adds a penalty for overly complex models (large parameters) to the loss function, encouraging simpler models that generalize better to new data.
Explain the bias-variance tradeoff conceptually.
Bias: error from oversimplified assumptions (underfitting). Variance: error from sensitivity to training data fluctuations (overfitting). Optimal models balance both to minimize total error.
What is cross-validation and what problem does it solve?
Dividing data into k folds, training k times (each fold as test once) gives a robust estimate of model performance without wasting data, reducing variance in performance estimates.
Make your own Machine Learning study set
Flashcards for related topics
Studying Machine Learning to build with AI? MindloomHQ turns it into real skills — structured courses, agent projects, and certificates.
Explore MindloomHQ →