Consider a variant of the Winnow algorithm, which sets Wi,t+1 ←4 Wi,t when it makes a mistake on positive examples and sets Wit+1 ← Wit / 8 when it makes a mistake on negative examples. What is the tightest upper bound for the number of algorithm mistakes for learning monotone disjunctions? (Ignore lower-order terms and only indicate the largest asymptotic term.) a. (44k/9) log n b. (31k/14) log n c. (k/3) log n d. None of the other answers are correct. e. (28k/9) log n
Consider a variant of the Winnow algorithm, which sets Wi,t+1 ←4 Wi,t when it makes a mistake on positive examples and sets Wit+1 ← Wit / 8 when it makes a mistake on negative examples. What is the tightest upper bound for the number of algorithm mistakes for learning monotone disjunctions? (Ignore lower-order terms and only indicate the largest asymptotic term.) a. (44k/9) log n b. (31k/14) log n c. (k/3) log n d. None of the other answers are correct. e. (28k/9) log n
Related questions
Question
Expert Solution
This question has been solved!
Explore an expertly crafted, step-by-step solution for a thorough understanding of key concepts.
Step by step
Solved in 2 steps