Neural networks

A core subset of machine learning (ANNs), neural networks mimic the human brain's interconnected neurons to process complex data: key applications include image and speech recognition.

Neural networks operate on a layered architecture (input, hidden, output) of nodes (artificial neurons) with weighted connections. The system learns via backpropagation: processing massive datasets and iteratively adjusting those weights to minimize prediction error. This adaptive, non-linear processing capability drives modern deep learning applications, enabling high-accuracy solutions in critical areas like large language models (GPT, BERT) and autonomous vehicle systems.

https://cloud.google.com/discover/what-is-a-neural-network