🤖

Machine Learning: Algorithms & Metrics

75 cards · 🇺🇸 English

Core ML algorithms and the metrics used to evaluate them.

Study all 75 cards for free

Flashcards, quizzes, pixel-art games and a daily review that remembers for you.

Study this deck free

Try the first 10 cards

1 / 10
Question

Machine learning

Knew itLearning

Tap the card for the answer · swipe right if you knew it, left if not

Already have an account? Sign in

Cards

  1. 1Machine learning
    A field of AI concerned with statistical algorithms that can learn from data and generalize to unseen data without explicit instructions
  2. 2Tom Mitchell's definition of machine learning
    A program learns from experience E with respect to task T and performance measure P if its performance at T, measured by P, improves with E
  3. 3Three main types of machine learning
    Supervised learning, unsupervised learning, and reinforcement learning
  4. 4Supervised learning
    Learning from labelled training data where inputs are paired with desired outputs
  5. 5Unsupervised learning
    Learning from unlabelled data to find patterns, structures, or relationships without predefined outputs
  6. 6Reinforcement learning
    Learning by interacting with an environment, receiving rewards or penalties, and maximizing cumulative reward
  7. 7Two main supervised learning tasks
    Classification (predicting discrete labels) and regression (predicting continuous values)
  8. 8Classification
    Predicting a discrete category or label from input features (e.g., spam/not spam)
  9. 9Regression
    Predicting a continuous numerical value from input features (e.g., house price)
  10. 10Clustering
    An unsupervised learning task that groups similar data points together
  11. 11Dimensionality reduction
    Reducing the number of features while preserving important information
  12. 12Training set
    The data used to train the model
  13. 13Test set
    The data used to evaluate the model's performance on unseen data
  14. 14Validation set
    Data used to tune hyperparameters and select models during training
  15. 15Overfitting
    When a model learns the training data too well, including noise, and performs poorly on new data
  16. 16Underfitting
    When a model is too simple to capture the underlying patterns in the data
  17. 17Bias-variance tradeoff
    High bias leads to underfitting; high variance leads to overfitting. Need to balance both
  18. 18Feature
    An individual measurable property or characteristic used as input to a model
  19. 19Label
    The output or target variable that the model is trying to predict
  20. 20Feature engineering
    Creating, selecting, or transforming features to improve model performance
  21. 21Confusion matrix
    A table showing true positives, true negatives, false positives, and false negatives for classification
  22. 22Accuracy
    The proportion of correct predictions: (TP + TN) / (TP + TN + FP + FN)
  23. 23Precision
    The proportion of positive predictions that are correct: TP / (TP + FP)
  24. 24Recall (sensitivity)
    The proportion of actual positives correctly identified: TP / (TP + FN)
  25. 25F1 score
    The harmonic mean of precision and recall: 2 * (precision * recall) / (precision + recall)
  26. 26False positive
    When the model predicts positive but the actual label is negative (Type I error)
  27. 27False negative
    When the model predicts negative but the actual label is positive (Type II error)
  28. 28ROC curve
    Receiver Operating Characteristic curve - plots true positive rate vs false positive rate at various thresholds
  29. 29AUC
    Area Under the ROC Curve - a single metric summarizing model performance (1.0 is perfect)
  30. 30Cross-validation
    A technique to assess model performance by splitting data into multiple folds and training/testing on different combinations

And 45 more cards. Study the whole deck free

Articles about this deck

More Programming decks

See all
Study all 75 cards free