Why do convolutional neural networks work well for images?
Deep learning · Intermediate 2
Short answer
Convolutions slide small learned filters across the image, so the same weights detect a pattern such as an edge anywhere in the picture. That weight sharing needs far fewer parameters than a fully connected layer and builds in translation equivariance. Stacked layers combine edges into textures, parts and objects, and pooling or strides shrink the image as features grow richer.
Deeper answer
A 224 by 224 colour image has about 150,000 input values; a dense layer on that is huge and ignores that nearby pixels relate. Convolution uses local receptive fields: each output looks only at a small patch, and stacking layers widens what deeper units see.
Modern practice rarely trains a CNN from scratch: you take a pretrained backbone such as a ResNet or EfficientNet and fine-tune it. Vision transformers now compete with CNNs on large datasets, but CNNs remain strong and efficient, especially on smaller data and edge devices.
What a strong answer covers
- Small filters slide across the image
- Weight sharing means fewer parameters
- Detects patterns anywhere (translation)
- Layers build up from edges to objects
- Pooling or strides reduce size
Answer it out loud in the mock interview about 1.5 min
Follow-ups to expect
- What does a 1 by 1 convolution do?
- When might a vision transformer beat a CNN?
Common mistakes
- Saying CNNs are rotation invariant by design
- Ignoring transfer learning
Learn it
- Explain backpropagation in simple terms.
- What is the Chinchilla scaling law, and why do labs now train models far beyond it?
- How would you estimate the compute and cost to train a language model from scratch?
- Which fine-tuning settings matter most, and how do you choose them?
- Your neural network overfits. What do you try?
- What is transfer learning, and when would you use it?
- What are vanishing and exploding gradients, and how are they addressed?
- What is model merging, and when would you use it?