elman

Elman is a surname and given name used in several cultures. It can refer to people bearing the name and, in some contexts, to concepts named after such individuals.

In science and artificial intelligence, the Elman network is a type of simple recurrent neural network named after Jeffrey L. Elman, an American linguist and cognitive scientist who introduced it in 1990. The architecture consists of an input layer, a hidden layer, an output layer, and a context layer that stores the previous hidden state, giving the network short-term memory.

The network processes sequences by incorporating information from prior time steps, enabling it to learn temporal dependencies. Training is typically performed by backpropagation through time. The Elman network played a foundational role in early recurrent architectures and helped motivate later developments such as long short-term memory models. It remains a standard example in discussions of context in neural networks.

Beyond this neural network, Elman appears as a surname for various individuals across fields and regions, reflecting its broader use in personal naming.