Вопрос
Explain what a convolutional neural network is to senior engineers who do not specialize in ML. Keep it accurate but accessible.
Ответить самому
Сначала сформулируйте ответ как на собеседовании, затем откройте разбор и оцените себя.
Короткий ответ
A CNN applies the same small learned filter across an image, detecting local patterns efficiently and composing them into higher-level features.
Полный разбор
A regular fully connected layer would connect every pixel to every output, which is expensive and ignores the fact that nearby pixels form local patterns. A convolutional layer uses a small learned filter, for example 3x3 or 5x5, and slides it across the image.
At each location, the filter computes a weighted sum over a local neighborhood. The same filter weights are reused at all positions, so the model learns a pattern such as an edge, texture or shape fragment and can detect it anywhere in the image. Multiple filters learn multiple pattern types.
Deeper layers compose local patterns into higher-level concepts. Early layers may detect edges and colors; later layers can represent parts and objects. Pooling or striding can reduce spatial resolution and increase the receptive field. The key ideas are locality, weight sharing and hierarchical feature extraction.
Теория
Convolutional layers exploit image locality and translation sharing to reduce parameters and learn reusable visual patterns.
Типичные ошибки
- Explain CNN as just “splitting the image into pieces”.
- Forget weight sharing across positions.
- Ignore that filters are learned, not manually fixed.
- Overcomplicate the explanation with continuous convolution math before giving intuition.
Как отвечать на собеседовании
- Use “small learned filter sliding across the image” as the core phrase.
- Mention locality and weight sharing.