In the slides’ 2 by 2 max pooling example with stride 2, the main effect is that the feature map becomes:

Question

BlackTom AI · Accepted Answer

2x2 max pooling with stride 2 reduces the spatial dimensions by roughly a factor of two in each direction, so the feature map becomes smaller.

1) Relevant concepts
- Max pooling: a down-sampling operation that replaces each local window (here 2x2) with its maximum value.
- Stride: the step size by which the pooling window moves. With stride 2 and a 2x2 window, windows are non-overlapping and cover the input in 2×2 blocks.
- Output size formula (no padding): output = floor((input_size - pool_size)/stride) + 1. For pool_size = stride = 2 this simplifies to about floor(input_size/2).

2) Step-by-step reasoning
- Each 2x2 block of the input is collapsed to one value (the maximum) because of max pooling.
- Because the stride equals the window size (2), blocks do not overlap and each block maps to exactly one output location.
- Thus the number of output locations along each spatial dimension is approximately half the input’s number of locations.

3) Example calculations
- If input is 8×8: output = floor((8 - 2)/2) + 1 = floor(6/2) + 1 = 3 + 1 = 4 → 4×4 (exactly half: 8/2 = 4).
- If input is 7×7: output = floor((7 - 2)/2) + 1 = floor(5/2) + 1 = 2 + 1 = 3 → 3×3 (floor(7/2) = 3).

4) Why the correct answer is correct
- Max pooling with window 2×2 and stride 2 performs down-sampling: each 2×2 area becomes one value, so the feature map’s spatial resolution decreases (gets smaller) while keeping the strongest activation per block.

5) Why other options are incorrect
- "Larger, because pooling expands the map": pooling does not create more spatial locations; it reduces them.
- "The same size, because stride does not matter": stride directly affects output size; stride 2 reduces it.
- "A text sequence, because pooling tokenizes data": pooling operates on spatial/feature maps, not converting maps into textual sequences.

6) Conclusion
The feature map becomes smaller, because pooling down-samples the map — matching the provided correct answer.

In the slides’ 2 by 2 max pooling example with stride 2, the main effect is that the feature map becomes:单项选择题

类似问题

Max pooling works by:

Convolutional Neural Networks effectively use pooling techniques for dimensionality reduction of feature maps. Apply MAX-POOLING to the example assuming a 2x2 filter and a stride of 2. What is the resulting value of x in the solution? Note: This will be an integer value.

How many learnable parameters are present in a max-pool layer of size 8 x 8 with stride 4?

In CNNs, max pooling:

What is the primary purpose of max pooling in a CNN?

In which security measure will you find anti-virus software?

What are user permissions?

Which security measure will stop someone reading your messages if they find your smart phone?

更多留学生实用工具

智能学习助手

风格化写作助手

论文查重助手

文献引用助手

课堂转译助手

课堂笔记助手

Quiz搜索助手

学校历年真题

智能刷题助手

智能匹配练习

希望你的学习变得更简单