Skip to content
ml-recall

#convolution-as-matrix

2 questions from Segmentation

Timed test

2 questions · 2 min · score and explanations at the end

Take a test
Q 1 / 2Score 0 / 0

Question 1: Short answer

mediumAI-checkedSegmentation

X = torch.arange(9.0).reshape(3, 3) and K = torch.tensor([[1.0, 2.0], [3.0, 4.0]]). Cross-correlation gives Y = corr2d(X, K) = [[27, 37], [57, 67]]. A transposed convolution with the same kernel then gives Z = trans_conv(Y, K), which has the same 3×33 \times 3 shape as X. What is Z[1, 1], the center entry (where X[1, 1] = 4)?

a whole number