Bioinspired Neural Network Model Significantly Enhances Memory Performance

Bioinspired neural network model can store significantly more memories

Researchers have developed a model that is inspired by recent discoveries in biology and shows improved memory performance. Modifying a classic neural network was the key to achieving this.

Computer models are crucial in the study of how the brain makes and retains memories, and other complex information. Constructing such models can be a delicate process. The complex interplay between electrical and biochemical signalling, and the web of connections among neurons and other types of cells, is what creates the foundation for memory formation. The limited understanding of brain biology has made it difficult to encode the complex biology into a computer-based model that can be used for further research.

By incorporating insights from the biological world, researchers at Okinawa Institute of Science and Technology have improved a widely used computer model of memory known as a Hopfield Network. Researchers at the Okinawa Institute of Science and Technology (OIST) have improved a widely used computer model of memory, known as a Hopfield network, by incorporating insights from biology.

Source:

Bioinspired Neural Network Model Can Store Significantly More Memories