Knowledge Graph — Coursera Notes › Academic disciplines › Computer Science / Information Technology › Artificial Intelligence › Deep Learning
CNN
concept · part of Deep Learning
A convolutional neural network (CNN) is a neural network architecture with convolutional layers followed by pooling and fully connected layers, suited for image classification. It uses conv and pooling layers, and can be implemented with multiple convolutional and fully connected layers, such as on the CIFAR-10 dataset. CNNs are specialized for grid-like data such as images, using convolutional layers to detect patterns like edges and pooling layers to reduce dimensions. They are used in image classification, object detection, and video analysis.
CNNs are also used for recognizing handwritten digits from MNIST and road signs in self-driving cars.
CNNs automatically learn spatial hierarchies of features from images and are a key choice in model development for computer vision tasks.
Inside CNN (2)
- Convolutional layer — Convolutional layers apply filters (kernels) that slide over input data to produce feature maps, detecting patterns like edges and textures.
- Pooling layer — Pooling layers reduce spatial dimensions of data, decreasing computational load and focusing on important features.
Connections
- Used for CIFAR-10
- Related to Caffe
- Related to Deep Learning
This is the text view of an interactive 3D knowledge graph — open this page with JavaScript enabled to explore it visually.