A convolutional neural network (CNN) attempts to recognize objects in images but fails to generalize to rotated versions of those objects. Which mechanism explains this failure?

A)Insufficient regularization of feature maps

B)Lack of rotation-invariant feature encoding✓

C)Overfitting on specific training examples

D)Suboptimal learning rate during training

💡 Explanation

The CNN's failure to generalize rotated objects arises because it lacks rotation-invariant feature encoding; it learns features specific to orientations rather than recognizing the object regardless of its orientation. Therefore, the CNN cannot extract the same features when the object is rotated, rather than because of overfitting or regularization issues.

🏆 Up to £1,000 monthly prize pool

Ready for the live challenge? Join the next global round now.
*Terms apply. Skill-based competition.

⚡ Enter Arena

A convolutional neural network (CNN) attempts to recognize objects in images but fails to generalize to rotated versions of those objects. Which mechanism explains this failure?

💡 Explanation

Related Questions