Essay

Explain the relationship between beam search approximations and scoring function optimization during inference.

Question: Discuss why approximate search algorithms such as beam search are used during inference to optimize a scoring function like Score_A(S), and explain why these algorithms are not guaranteed to identify the output that maximizes this score.

Sample answer: Beam search is an approximate search algorithm used during inference to find the value of S that optimizes (maximizes) a scoring function, Score_A(S). To manage search complexity, beam search keeps only the top K candidates at each step of the search process. Because it is an approximate method that discards other potential candidates rather than performing an exhaustive search, it is not guaranteed to find the value of S that maximizes Score_A(S).

Key points:

  • Beam search is an approximate search algorithm used to find the value of S that maximizes Score_A(S).
  • Beam search keeps only the top K candidates during the search process.
  • Approximate search algorithms are not guaranteed to find the value of S that maximizes Score_A(S).

Rubric: The response must contain: 1) Identification of beam search as an approximate search algorithm used to optimize/maximize Score_A(S). 2) Explanation that beam search keeps only the top K candidates during the search process. 3) Statement that approximate search algorithms like beam search do not guarantee finding the output that maximizes the scoring function.

0

1

Updated 2026-05-26

Contributors are:

Who are from:

Tags

Machine Learning

Deep Learning

Supervised Learning

Dive into Deep Learning @ D2L

Data Science

Machine Learning Strategy