Optimization DY0-001 Practice Questions (Page 2)

Questions 6–10

1. 6foundation · easy

   What type of constraint is represented by the requirement that a variable must be an integer?

   • AInequality constraint
   • BIntegrality constraint
   • CEquality constraint
   • DBound constraint
   Reveal Solution Discussion Select an answer first
2. 7foundation · easy

   Which of the following best describes the role of an objective function in an optimization problem?

   • AIt defines the constraints that limit the possible solutions
   • BIt determines the learning rate for gradient-based methods
   • CIt stores the historical performance of previous model iterations
   • DIt quantifies the quality of a candidate solution that the optimizer seeks to improve
   Reveal Solution Discussion Select an answer first
4. 8foundation · easy

   What is a key characteristic of a constrained optimization problem?

   • AThere are no limits on the values the decision variables can take
   • BThe objective function is always linear
   • CThe solution must satisfy one or more equality or inequality restrictions
   • DThe optimal solution is always found at a stationary point of the objective function
   Reveal Solution Discussion Select an answer first
5. 9foundation · easy

   In data science, what is the primary purpose of optimization?

   • ATo ensure that a model's predictions are always correct on training data
   • BTo automatically label unlabeled data for supervised learning
   • CTo reduce the size of a dataset by removing redundant features
   • DTo find the best set of parameters that minimizes or maximizes an objective function
   Reveal Solution Discussion Select an answer first
6. 10application · medium

   A data scientist is fitting a curve to a set of data points. The goal is to find the parameters of the curve that minimize the sum of squared errors. There are no restrictions on the parameter values. Which optimization technique is most appropriate for this task?

   • ASimulated annealing, because it is the only technique that can handle unconstrained problems.
   • BLinear programming, because the sum of squared errors is a linear function.
   • CInteger programming, because the parameters must be whole numbers.
   • DGradient descent, because it is an iterative method that can find the minimum of a differentiable function without constraints.
   Reveal Solution Discussion Select an answer first