CNN Cheat Sheet 2026

The 30 highest-yield CNN facts, distilled from real exam questions. Print it, save it as a PDF, or study it here — free, no sign-up.

1. Which layer type is typically removed when adapting a pretrained CNN for a new classification task? → Fully connected output layer
2. What is the 'dying ReLU' problem? → Neurons with ReLU permanently output zero because their weights push inputs negative
3. What is the purpose of transposed convolutions (deconvolutions) in segmentation networks? → To learn to upsample feature maps back to a higher spatial resolution
4. The Leaky ReLU activation function was designed to address which limitation of standard ReLU? → Zero gradient for negative inputs leading to dead neurons
5. Which of the following regularization techniques involves halting the training process when the model's performance on a validation set stops improving? → Early Stopping
6. What problem does Non-Maximum Suppression (NMS) solve in object detection? → Multiple overlapping predictions for the same object
7. During CNN training, which data augmentation technique generates new training samples by combining two images and their labels linearly? → Mixup
8. What is the effect of applying Gaussian blur augmentation to CNN training images? → It simulates out-of-focus conditions, making the model robust to image blurriness
9. During CNN training, which regularization method randomly drops entire feature maps rather than individual neurons? → DropBlock
10. What is the depth of the winning ResNet model in the ILSVRC 2015 competition? → 152 layers
11. What effect does L1 regularization have on CNN weight vectors compared to L2? → It produces sparse weights with many zeros
12. A CNN trained on 10,000 images achieves 99% training accuracy but only 62% validation accuracy. Which symptom does this describe? → Overfitting
13. Which PyTorch library is most commonly used for applying data augmentation transforms to image datasets? → torchvision.transforms
14. Global Average Pooling (GAP) is frequently used at the end of a CNN to replace which traditional layer? → Fully connected (dense) layer
15. What does AutoAugment do in the context of CNN training? → Searches for an optimal augmentation policy using reinforcement learning
16. What is 'negative transfer' in the context of transfer learning with CNNs? → When pretraining hurts performance on the target task
17. Which activation function is most commonly used in the hidden layers of modern deep CNNs? → ReLU
18. Which Python library function is commonly used to extract intermediate layer outputs for feature map visualization in Keras? → Model(inputs, intermediate_layer.output)
19. Which of the following was a key innovation introduced in the AlexNet architecture that distinguished it from its predecessor, LeNet-5? → The use of the Rectified Linear Unit (ReLU) activation function.
20. What loss function is most commonly used as the primary training objective in semantic segmentation CNNs? → Pixel-wise Cross-Entropy Loss
21. Dialysate glucose must be heat sterilized at a low pH to → Decrease generation of glucose degradation products
22. What is the receptive field of a neuron in a CNN? → The region of the input image that influences that neuron's activation
23. Which of the following is NOT a common technique used to combat overfitting in a CNN? → Increasing the number of epochs indefinitely.
24. What is the key structural unit introduced by GoogLeNet (Inception v1)? → Inception module
25. How does a transposed convolution (sometimes called deconvolution) differ from a standard convolution in terms of spatial output? → It increases spatial dimensions, making it useful for upsampling
26. When visualizing early-layer feature maps, what kind of patterns are typically detected? → Low-level patterns like edges, colors, and textures
27. In the context of a convolutional operation, what does the 'stride' parameter define? → The number of pixels by which the filter slides over the input at each step.
28. Which quantity represents the number of output feature maps produced by a convolutional layer? → Number of filters
29. Which architecture introduced the concept of 'network-in-network' using micro neural networks at each convolutional step? → Network-in-Network (NIN)
30. A deep CNN with many layers is tested without any regularization on a small dataset. Which problem is MOST likely? → Overfitting due to excess model capacity