Neural Network Architectures

Different tasks demand different network structures, each tuned to a data type:

• Convolutional Neural Networks — built for images, using sliding filters to spot patterns anywhere in the frame; they power facial recognition and medical imaging
• Recurrent Neural Networks — built for sequences like text and speech, keeping a memory of previous inputs for translation and voice recognition

Transformers are the architecture behind ChatGPT and modern AI. They process an entire sequence at once instead of one element at a time, which makes them dramatically faster and explains why language models improved so quickly after 2017.