Machine Learning Basic Terminologies
37 Questions
5 Views

Choose a study mode

Play Quiz
Study Flashcards
Spaced Repetition
Chat to lesson

Podcast

Play an AI-generated podcast conversation about this lesson

Questions and Answers

What does regression primarily predict?

  • A binary outcome
  • A continuous number (correct)
  • A grouping of instances
  • A category of labels
  • Which of the following best describes a label in machine learning?

  • The variables used to predict an outcome
  • The final choice predicted by the model (correct)
  • The input data for the algorithm
  • The technique used for clustering
  • What type of data consists of distinct categories without any natural ordering?

  • Ordinal data
  • Numeric data
  • Continuous data
  • Categorical data (correct)
  • Which machine learning model is not typically associated with classification tasks?

    <p>K-means clustering</p> Signup and view all the answers

    Which of the following is true about features in machine learning?

    <p>They describe the input data.</p> Signup and view all the answers

    What distinguishes ordinal data from categorical data?

    <p>Ordinal data has a natural ordering.</p> Signup and view all the answers

    What is an example of a feature in a spam detector?

    <p>The sender's address</p> Signup and view all the answers

    What is typically used to encode categorical data?

    <p>Numeric values</p> Signup and view all the answers

    Which type of feature represents the loan amount in credit approval applications?

    <p>Numeric</p> Signup and view all the answers

    What is the primary purpose of applying linear regression?

    <p>To predict relationships between variables</p> Signup and view all the answers

    In the context of handwritten digit recognition, which type of feature categorizes pixels based on color?

    <p>Categorical</p> Signup and view all the answers

    What does the term 'hard feature engineering process' refer to in machine learning?

    <p>The conversion of raw data into a structured format</p> Signup and view all the answers

    In the equation for linear regression, what does 'b' represent?

    <p>The y-intercept</p> Signup and view all the answers

    Which label is an output option for credit applications?

    <p>Approve application</p> Signup and view all the answers

    What characteristic makes ordinal features distinct in machine learning?

    <p>They have a natural order</p> Signup and view all the answers

    What does 'y’' represent in the adjusted linear regression equation?

    <p>The predicted label</p> Signup and view all the answers

    What is the primary purpose of feature engineering in machine learning?

    <p>To extract and transform variables for training and prediction</p> Signup and view all the answers

    During which phase of a model's life is it trained using labeled examples?

    <p>Training Phase</p> Signup and view all the answers

    What are labeled examples primarily used for in machine learning?

    <p>Training the model</p> Signup and view all the answers

    What does a model define in the context of machine learning?

    <p>The relationship between features and a label</p> Signup and view all the answers

    What is the purpose of taking a random sample when estimating a population characteristic, such as average height?

    <p>To provide a close approximation of the characteristic</p> Signup and view all the answers

    What best describes the concept of convergence in the context of gradient descent?

    <p>It indicates when a model reaches a stable state and stops improving its predictions.</p> Signup and view all the answers

    Which characteristic is NOT a property of a gradient in the context of gradient descent?

    <p>Weight</p> Signup and view all the answers

    What role does the learning rate play in the gradient descent algorithm?

    <p>It sets the step size at each iteration while moving towards a minimum of the loss function.</p> Signup and view all the answers

    In gradient descent, which of the following statements is true about the update step?

    <p>It adds a fraction of the gradient’s magnitude multiplied by the learning rate to the previous point.</p> Signup and view all the answers

    What type of plots do regression problems yield when considering loss versus weight?

    <p>Convex plots showing a clear minimum.</p> Signup and view all the answers

    Which statement about the steps in the gradient descent algorithm is correct?

    <p>The algorithm iteratively improves until convergence is achieved.</p> Signup and view all the answers

    What does the phrase 'greedy algorithm' imply in the context of taking steps towards a minimum in gradient descent?

    <p>It means only the current gradient is considered for the next step decision.</p> Signup and view all the answers

    What is typically indicated by a 'smiley face' graph in the context of convex functions?

    <p>The establishment of an optimal point at the bottom.</p> Signup and view all the answers

    What is the purpose of the learning rate in gradient descent algorithms?

    <p>To multiply the gradient to adjust step sizes</p> Signup and view all the answers

    Which gradient descent method processes training data in small batches or individually?

    <p>Stochastic Gradient Descent (SGD)</p> Signup and view all the answers

    In logistic regression, what type of outcome does it predict?

    <p>Probability of a binary event occurring</p> Signup and view all the answers

    Which statement describes Batch Gradient Descent?

    <p>Involves averaging gradients from the entire dataset</p> Signup and view all the answers

    What condition represents Class 1 in logistic regression's decision boundary equation?

    <p>0 &lt; w2x2 + w1x1 + w0</p> Signup and view all the answers

    What is a characteristic of Mini-Batch Gradient Descent?

    <p>Combines features of both Batch and Stochastic methods</p> Signup and view all the answers

    What is the loss function used in the gradient descent approach mentioned?

    <p>Mean Squared Error</p> Signup and view all the answers

    What indicates a limitation of logistic regression's output?

    <p>It can only predict two outcomes</p> Signup and view all the answers

    Study Notes

    Machine Learning Basic Terminologies

    • Labels: Output that's predicted in a classification or regression task. Often denoted with the variable 'y'.
    • Features: Input variables that describe the data. Denoted as x1, x2,...xn.
    • Examples: A specific data instance, denoted as 'x' and often has an associated label. Examples can be labeled or unlabeled.
    • Models: Define a relationship between features and labels. Have two phases – training and testing/inference.

    Types of Features

    • Categorical: Information that can be put into categories. Example: Red, blue, brown, yellow.
    • Ordinal: Information that has a natural order. Example: Poor, satisfactory, good, excellent.
    • Numeric: Can be expressed as an integer or real value. Example: Height, weight, temperature.

    Feature Engineering

    • Feature Engineering: Extracting and transforming variables from raw data to use for training and predictions.

    Model Building: Gradient Descent

    • Gradient Descent: An optimization algorithm that updates model parameters iteratively to minimize loss.
    • Steps: Pick a random starting point for the model parameters, calculate the gradient of the loss function, update parameters, repeat steps 2 and 3 until convergence.
    • Convergence: When the model stops improving its predictions.
    • Learning Rate: Parameter that controls the step size at each iteration of gradient descent.

    Types of Gradient Descent

    • Batch Gradient Descent: Uses the entire training dataset for each iteration. Can be slow for large datasets.
    • Stochastic Gradient Descent (SGD): Processes data in small batches or individual data points. Noisy but fast.
    • Mini-Batch Gradient Descent: Compromise between Batch Gradient Descent and SGD. Uses a batch of a fixed size.

    Logistic Regression

    • What it is: Used for binary classification tasks. Predicts the probability of a binary event occurring (e.g., yes/no, 0/1, true/false).
    • Input: Features values.
    • Output: Probability from 0 to 1.
    • Linear Classifier: The decision boundary is represented by a linear equation.
    • Example: Predicting whether a student will pass or fail an exam based on the number of hours they study.

    Linear Regression

    • What it is: Attempts to model the relationship between two variables by fitting a linear equation.
    • Example: Predicting temperature based on the number of cricket chirps.
    • Equation: y' = w1x1 + w0
      • y': predicted label
      • w1: weight of feature 1
      • w0: bias (intercept)
      • x1: input feature value

    Studying That Suits You

    Use AI to generate personalized quizzes and flashcards to suit your learning preferences.

    Quiz Team

    Related Documents

    Description

    This quiz covers essential terminologies in machine learning, including key concepts like labels, features, and models. Understand different types of features such as categorical, ordinal, and numeric, as well as the importance of feature engineering in model building. Test your knowledge and deepen your understanding of these foundational concepts.

    More Like This

    Use Quizgecko on...
    Browser
    Browser